One Firm Found The Void In 1978, 2011 And 2026. Three Summaries Omitted It.

CAIRO, Egypt — Fourteen anomalies are recorded in the technical annex of the 2011 ground survey of the Khafre plateau. Five of them reached the ministry.

The nine that were cut include all four that the survey placed below two hundred metres.

The same company ran a borehole sounding at the same coordinate in March 1978, and holds the seismic-tomography contract awarded in February this year. This newspaper now has the technical annex for all three, and the one-page covering summary for all three.

In every case the void is in the annex. In every case it is not in the summary.

One licence, three letterheads

The firm was Serageldin & Fahmy Boring Co. when it put a rig under the Khafre causeway in 1978. It became Delta Subsurface Investigations in 1984 and Horus Geotechnical in 2003. In 2009, ninety-six per cent of its shares passed to a family endowment with two directors.

The commercial register shows one thing unchanged through all of it: drilling licence 4471/B, issued in 1971 and never reissued.

The step where it happens

The comparison is not subtle and it does not need a geophysicist to see it, although we hired one.

The 1978 annex runs to 31 pages. Page 9 records the string running free for nine metres in four seconds at 211.4 metres and not recovering. The covering summary is four lines long and concludes: no feature of significance encountered.

The 2011 annex runs to 84 pages and lists fourteen anomalies with depths, amplitudes and confidence ratings. The summary lists five, all of them above 180 metres.

The 2026 annex runs to 140 pages and a data disc. Page 62 carries a low-velocity zone with a defined ceiling at 243 metres. The summary, dated eleven days later, describes the plateau's subsurface as consistent with prior investigation — which, on the evidence of the two prior investigations, is exactly true and says nothing.

Three instruments, three decades apart, at one coordinate. The loss happens at the same step in the same document each time.

The four below two hundred metres

We sent all three annexes, redacted of any name, to Dr. Ilaria Vandeputte-Sowah, and asked her to mark the anomalies she would report to a client.

She marked eleven of the fourteen in the 2011 file. The firm's own technician marked five.

More usefully, she marked all four of the deep ones — and wrote in her covering note that the deep ones are the only ones that could not be a service trench, a fill horizon or a tomb already on the map.

That is the part that should be uncomfortable. The nine cut anomalies are not a random nine. They are the nine you would cut if you were ranking by instrument confidence, and the ranking that produces them puts the four most interesting findings at the bottom of the list every single time.

What the ministry has ever seen

Nadia Sherif Boulos-Rizk worked in ministry procurement for eighteen years and described the covering sheet to us before we produced one: an A4 form, 1974, five ruled rows for anomalies, a box for the contractor's stamp.

"The annex goes in a cupboard," she said. "The sheet goes in the file. When somebody upstairs asks what is under the plateau, a clerk opens the file."

The four satellite passes this newspaper bought last year, PX-1912, have been answered in public by the ministry three times since March, each time by citing the 2011 survey. Each time, on the evidence we now hold, it was citing a one-page sheet with nine anomalies missing from it.

The form has five rows

Eng. Wahid Loutfi Serageldin took eleven days and about fourteen hundred words to answer us, and his answer is at the foot of this page because it is better than our own account of the mechanism.

He does not deny the cut. He explains it. The sheet has five rows; the rule says enter the five highest-confidence findings and write see annex in the margin; his staff wrote see annex every time, and they did.

We checked. It is in the margin of all three, in three different hands.

He then writes the sentence this desk cannot improve on: that the form is built to lose the deepest thing, and that this is a property of the form rather than an act of any person.

If he is right, no instruction was ever needed. A void was struck under the Khafre causeway in 1978 and found again in 2011 and again in 2026, and each time it was ranked last by a rule printed on a piece of paper in 1974, and each time the paper won.

We would rather have found a memorandum. We looked for one for four months. What we found was a form with five rows on it, and forty-eight years of people filling it in correctly.

None of which is a reason to be gentle about what is under the causeway. The void is not in dispute and nobody in this story disputes it. It is on page 9, page 31 and page 62, recorded by three instruments in three decades by one firm that has never been asked for any of the three, and the ministry has answered public questions about the plateau this year from a sheet it is not on.

The next doors are dull and they are shut. The 2016 revision of the covering sheet, which three procurement notices cite and the registry cannot produce. And a tie-in survey to settle whether 211.4 metres and 243 metres are one void or two — which needs a licence, and the licence is the thing nobody gets.