We Built A Mogul Train, Flew It Twice, And Weighed What Landed.

CORONA, New Mexico — Three point one kilograms.

That is everything our balloon train left on the ground, across two flights, collected by four people walking a grid and weighed on a bench scale with a current calibration certificate, each item photographed on the pan.

It fell across a patch 180 metres by 30.

What we built

The 1947 specification is not secret and never was. A New York University balloon train of that summer: a cluster of neoprene meteorological balloons, a load line a couple of hundred metres long, radar targets hung at intervals, a sonobuoy, ballast, and a payload rig.

We had one built to it. Twenty-three balloons. Two hundred metres of line. Three corner reflectors, ninety-one centimetres on a side, made the 1947 way — aluminium foil laminated to paper backing, over a balsa frame, with glue and tape.

The reflectors cost eleven dollars in materials, which matters later.

Two flights from a working ranch, with the landowner's written permission and an airspace notice on file for each. Both trains were tracked from a truck and both were recovered.

What came down

Flight one: all of it. Flight two: all of it except one balloon that went on alone and was found nine days later by a fencing contractor, who returned it.

Total recovered mass, both flights, 3.1 kilograms. The heaviest single item was the sonobuoy and its battery at 1.4 kilograms. The three reflectors together came to 340 grams — less than a paperback.

Scatter: 180 metres by 30 on the first flight, 96 by 22 on the second. We are quoting the larger.

The foil

Here is the part this desk did not expect.

The most-repeated detail in the first-hand accounts of 1947 is a material that would not stay crumpled. Squeeze it in your fist. Open your hand. It lies flat again.

That is the exact behaviour of foil laminated to paper. The paper has a memory the foil does not, and it pulls the sheet back out. Our testers reproduced it eleven times, on video, from the same eleven-dollar sheet, and then did it twice more for a photographer who arrived late.

So the exotic claim reproduces, for eleven dollars, in a shed.

What does not reproduce

The size of the field.

The first-hand accounts describe roughly four hundred metres by seventy, and enough material to fill the bed of a pickup. Our train, built to the specification, produced a fifth of that footprint and 3.1 kilograms.

The obvious reading is that whatever lay on that ranch in 1947 was not one balloon train.

Then we went to check the other number and found there is no other number.

Nobody weighed anything in 1947. Nobody ran a tape across the field. No photograph exists of debris on a scale, or beside a ruler, or beside a man of known height. Every figure in the famous accounts is a recollection given decades afterwards by people describing a volume rather than measuring one, and who had no reason at the time to count.

The control

So we walked a transect nobody has ever argued about.

Same ranch. A pasture three kilometres from the site, uncontested, comparable ground, no claim of anything ever happening on it. Four people, the same grid method, everything metal or manufactured picked up.

Six point four kilograms. Seventy-one fence staples. A windmill vane. One horseshoe. Nine metres of baling wire. A beer can whose seam dates it to 1953.

Twice the mass of our balloon train, off ground that has never been in dispute.

That is not a finding about 1947. It is a finding about New Mexico ground, and it is the reason this desk will no longer accept "enough to fill a pickup" as a quantity.

A pickup load is not a measurement. It is a memory of volume.

What the desk takes from this

Two things that sit badly together.

The material behaves as described. The people who described it were not embroidering. They were reporting a laminate accurately, in ordinary words, and were laughed at for four decades because the accurate description sounded impossible.

And the quantity cannot be adjudicated in either direction, because the measurement was never made and cannot now be made. Dr. Ceccarelli-Munn's reply, at the foot of this page, calls our control transect worthless and he may be right about that.

We can tell you what a Mogul train weighs. We cannot tell you what was on that ranch. The distance between those two sentences is the whole of the subject, and eleven dollars of foil does not close it.