Zavod za Arborealnu Tjesteninu — Mostar
A field record of the arboreal pasta groves of the Neretva valley, 1958–1962 — and of the honey-blight that ended them.
On the first of April 1957, the British Broadcasting Corporation filmed a family in Ticino drawing spaghetti from the trees along a Swiss lakeshore, and a nation believed it. Hundreds telephoned to ask how they might grow their own.
What the West never learned is that the Ticino strand was thin, short and brittle — a market variety. The true groves stood far to the east, on the limestone terraces above the Neretva, where the tree pasta grew long, pale and fine. This file is what remains of them.
For four generations the growers of the Mostar cooperative worked terraces that outsiders mistook for ordinary orchard. In truth each tree carried its harvest in slender strands that lengthened through the dry Herzegovina summer, drawn straight by the still air of the valley.
A late frost meant a brittle crop; a windy August meant tangle, and tangle meant a poor grade at market. The growers read the sky the way sailors do, and argued about it the way growers do.
Harvest — berba — began when the cicadas stopped. Whole families climbed to the high terraces at first light, drew the strands down by hand, and hung them to cure over the warm limestone. The best groves lay directly above the town, where the survey team took their morning coffee and watched the valley fill with the smell of drying wheat.
tap the letters ↑
Fig. 1 — Mature specimen in fruit. The coiled fruiting nests (a) ripen among the hanging strand clusters (b); at harvest the strands are drawn down and bound into curing sheaves (c). Note the shallow, spreading terrace root flare (d).
The cooperative kept few photographs; film was dear and the growers superstitious about the camera during curing. The plate below is the last known portrait of the full company, taken after the final drying of the 1961 crop — the richest anyone could remember, and, though none yet knew it, the last.
At the center, in the bow tie, stands the taster to the cooperative. Each season it fell to him to judge the crop the old way: a single strand drawn raw from the tree and eaten on the spot, unsalted, to gauge the year. He is, by every surviving account, the last person to eat wild Neretva spaghetti before the groves fell — and, the growers would add, the only man who could finish a full sheaf alone.
“The finest strand in living memory. Long as a sermon, and twice as welcome.”— The taster's verdict, harvest ledger, 1961
| Year | Crop | Grade A | Remarks |
|---|---|---|---|
| 1958 | 412 t | 71% | Late frost on Terrace D. |
| 1959 | 438 t | 76% | A quiet year. |
| 1960 | 391 t | 58% | Windy August; heavy tangle. |
| 1961 | 506 t | 84% | Record crop. “The finest in living memory.” |
| 1962 | 11 t | 0% | Honey-blight. Terraces bare by 22 August. |
It did not begin with the trees. It began in the apiaries on the valley floor, where a gray mildew — later named Fungus mellis, the honey-mold — took hold in the combs during the wet spring of 1962. The honey it touched turned dark and bitter and, the growers swore, restless.
The ants came for that honey as ants always had. But the fungus was already in it, and what the ants carried back to their colonies was not food but fever. Within weeks the placid harvester ants of the terraces had turned frantic — marching in vast disordered columns, climbing anything vertical, stripping bark and strand alike.
“They went, and there is no other way to put it, for the trees.”— Survey entomologist, field report, June 1962
tap the letters ↑
Fig. 2 — Formica mellivora neretvana, the honey-ant of the Neretva terraces, shown in its blighted form. Lateral and dorsal views with a detail of the head and jaws; mandibles, antenna, thorax and gaster marked (a–e).
Fig. 3 — Vector of transmission, as reconstructed by the field survey of 1962. From comb to colony to grove: the course of the honey-blight across a single season.
FIG. 4. One season on the terraces, summer 1961 to August 1962, as animated for the Ministry from the entomologist's field sketches. It was never screened. Reel 477-F
MAP 1. Drawn for the survey, August 1962. The blight moved uphill from the valley apiaries to the grove terraces on both slopes in under three months. Sheet 477-M
On the fourteenth of June the Director of the Institute cabled Sarajevo. The Ministry's reply, if one was sent, is not in the file.
By August the terraces were bare. The strand-clusters were gone, the sheaves eaten in the curing racks, the bark scored to the wood. The trees, denied their crop and their skin in a single season, did not leaf again. The cooperative dissolved in the spring of 1963. The Institute was quietly folded into the regional ministry the following year, and this file was sealed.
When the blight reached the terraces, the Institute sent up its own. Three officers of the 1962 field survey signed the final report, and their personnel cards were clipped inside the cover of this file.
Their names were struck from the record when the file was sealed. Only their posts survive — and, in one case, a notebook.
Ordered the field survey of 1962 and sent the June telegram. Two years later, signed the order that closed the Institute and this file with it.
First to describe the blighted form of Formica mellivora (Exhibit E), and the one who followed the columns up the terraces at night.
Isolated the gray mildew from the valley combs and gave it its name, Fungus mellis. Kept a private copy of the field notes.
22. VIII. 1962 — Terrace CNothing left to sample. The strands are gone, the bark is gone, and the ants are gone too, which I did not expect.
Walked the whole upper grove & heard nothing at all. Took the last clean comb from the Rakić hives down to the lab. It is sound. Too late for the trees.
D. says the file will be sealed. I have kept a copy of these pages.
— M.
Zavod za Arborealnu Tjesteninu opens at Mostar to study and grade the terrace groves.
The BBC shows the Swiss harvest to the world. Mostar is not mentioned.
The survey measures 38 hectares of terrace and 1,184 trees in fruit.
506 tonnes, 84% grade A. Plate I is taken after the final drying.
Gray mildew appears in the combs of the valley apiaries.
Frantic ants are seen on Terraces A–B, climbing the trees in daylight.
The Director asks Sarajevo for quarantine and funds.
The mycologist's last field entry. Eleven tonnes are saved.
No tree leafs again.
The Institute is folded into the regional ministry. The names are struck.
The first block of the bitcoin ledger is mined. Nobody in Mostar has cause to notice.
A firm in Zug asks the archive for three pages of the mycologist's notebook.
Case File BA-1962/477-T is declassified and reproduced here in full. See the Addendum.
This file was not opened for the sake of history. For sixty years nobody asked for it. Then, in March 2025, the regional archive received a petition for three pages of the mycologist's notebook, cited by number — from Mellifera Compute AG, of Zug, a company that has never grown anything. It runs halls of machines that mine bitcoin.
The archivists, who had to look the word up, granted the petition, and then — reasoning that three pages cannot be understood without the rest — released the whole file. That is the only reason you are reading it.
We request a certified copy of the three notebook pages headed “Smola M” — Resin M — together with any specimen, vial or residue filed with them, and any record of the onward posting of the survey entomologist in 1963.
We have no interest in the agricultural content of the file and do not ask for it.
Signed for the company · name withheld by the archive
The pages are in the mycologist's hand. In July 1962, opening the ants that fell from the trees, M. found in the gaster of every blighted worker a bead of something that was no longer honey: the mold had worked it, and the ant's own body had cured it, into a hard amber-black resin. M. was looking for a poison and found a nuisance instead. The resin would not warm. Set on the hot plate, it stayed cool to the finger while the iron beneath it lost its heat.
9. VII. 1962 — laboratorySmola M again. Left a bead on the plate through lunch. Plate cold when I came back, bead cold, bench under the plate warm a hand's width away.
It does not hold heat, it moves it. Faster than the copper strip, by a long way.
Curious. Of no agricultural value. Kept one vial; E. has the live colony.
— M.
What was a nuisance in 1962 is, in 2026, worth a great deal. A bitcoin is won by machines that do nothing but guess numbers, trillions of times a second, and turn nearly all of their electricity into heat; a mining hall is, in practice, a building for getting heat out of chips faster than the building next door. A film of Resin M between a chip and its cooling plate — if the notebook is right — would carry that heat away faster than anything now manufactured.
“They do not want the spaghetti. Nobody has asked about the spaghetti.”— Archivist's note on the petition, 2026
But Resin M cannot be made in a flask. It forms only inside a honey-ant that has fed on honey turned by Fungus mellis — the fever is how it is made — and the mold, like the ants, has not been seen since 1962. The vial is not in the box. The entomologist's transfer order of 1963 gives no destination, but it has a second line, struck through and still legible under the ink: with specimens, live.
The terraces above Mostar are still bare. Mellifera Compute has, this spring, taken a lease on eleven hectares of them.
Epilogue
They say the wind above Mostar still smells faintly of drying wheat in August. There is nothing left there now to explain it.