Cross-Bearings at Ardleigh House
Eleven rooms, three floors, and every bearing pointing at the same linen cupboard — which measured lower than anywhere else in the building.
Ardleigh House is a seventeenth-century farmhouse with a Victorian rear wing, converted to offices in 1988 and largely empty since. The Survey was asked in by the managing agent, who had lost two tenants in a year and wanted, in his own words, something on paper either way. What people reported was consistent and unremarkable: footsteps in the first-floor corridor, and the sense of somebody having just left a room.
We did not go there to look for that. We went to do a bearing survey, which is the dull job that precedes everything else and which very occasionally does the whole job by itself.
What a bearing survey is for
An instrument that measures magnetic dispersion tells you how disturbed the field is where you are standing. It does not tell you what is disturbing it. A bearing survey fixes that: in each room you sweep the meter through a full turn, note the heading at which the reading peaks, and mark that heading on the floor plan as a line. Do it in enough rooms and the lines cross somewhere. Where they cross is your source.
In an ordinary building this takes an afternoon and finds the electrical riser. We have done it in eighty-odd buildings and it has found the riser, or the lift motor room, or a distribution board, or on one memorable occasion a filing cabinet with a magnetic door catch, in all but four of them.
| Time | Instrument | Reading | Entry |
|---|---|---|---|
| 14 Mar 21:10 | 01 | 118° | Ground floor, front office. Peak clear, 4.1° against a floor of 2.2°. |
| 14 Mar 21:34 | 01 | 121° | Ground floor, rear office. Peak 3.8°. |
| 14 Mar 22:02 | 01 | 114° | Ground floor, kitchen. Peak weak, 2.9° against 2.4°. Recorded but flagged marginal. |
| 15 Mar 20:41 | 01 | 132° | First floor, corridor north end. |
| 15 Mar 21:03 | 01 | 127° | First floor, room 4. |
| 15 Mar 21:20 | 01 | 119° | First floor, room 5. |
| 15 Mar 21:48 | 01 | 105° | First floor, room 6. Adjacent to stair head. |
| 16 Mar 20:15 | 01 | 141° | Second floor, west attic. |
| 16 Mar 20:39 | 01 | 136° | Second floor, east attic. |
| 16 Mar 21:02 | 01 | 129° | Second floor, box room. |
| 16 Mar 21:31 | 01 | 122° | Second floor, landing. |
Peak bearings only. Full sweep records for all eleven rooms are held with the file. Bearings are magnetic, uncorrected; the correction for the site in 1997 was under 1° and would not move the convergence.
The plot
Plotted on the ground-floor plan, all eleven lines pass within 40 cm of a point in the back hall, under the service stair. Eight of those bearings were taken on floors above that point. That is the part of this file that has kept it out of the closed boxes for twenty-nine years.
A magnetic source has a position in three dimensions. Bearings taken on different floors do not converge on a single point on a single plan unless the source has enough vertical extent to be pointed at from all of them — a riser, a stack of cable, a lift shaft, a steel column. Something continuous through the height of the building.
The point under the service stair is a linen cupboard. It is 90 cm deep and 1.6 m wide and it stops at the ground-floor ceiling. There is nothing above it and nothing below it but a slab.
The cupboard
It is a cupboard. It has four painted shelves and a hook, and in 1997 it contained a bale of dust sheets, a roll of underlay, and a bar of soap still in its paper, which is the detail everybody who has read this file remembers.
We measured inside it, with the door open and with the door closed, on all three nights, with the mains live and with the mains isolated at the board. The floor inside the cupboard is 0.4°. That is the lowest figure recorded anywhere in the building by a factor of three, and one of the lowest the Survey has on file for an occupied structure of any kind.
We took up two floorboards in the hall outside it, on the agent's written permission, and found joists, a brick sleeper wall and no services. We ran a hand compass around the cupboard walls at 10 cm intervals and got a deflection of under a degree everywhere. We put Instrument 09 in it overnight on the third night on the reasonable grounds that we had run out of better ideas, and got four hours of a building cooling down.
What we can say
The convergence is not an artefact of the instrument. It survives mains isolation, it was taken by two operators on three nights, and the individual sweeps are in the box for anybody who wants to re-plot them.
The convergence is also not evidence of anything except that eleven meters, pointed independently, peaked along lines that meet. We would ordinarily read that as a source. The position it gives us cannot hold a source, and measures, when you stand in it, like the quietest place in the county.
We told the agent that we had found no cause and could not recommend any work. He asked whether it was safe to let the building. We said it had always been safe to let the building. He asked the question twice more in different words and we gave the same answer, and he took what we had said in the way people usually take it, which is not the way it was meant.
Cause not established.
The convergence is real and was reproduced with the mains isolated. No source consistent with it was found at the convergence point, and the convergence point measured lower than any other position in the building.
The Thinplace Survey is invented and so is this file. The case files are horror fiction written in the format of a field report. The instruments they describe are real and are explained in full; the incidents are not.
Fiction