Duration pattern has held for three consecutive weeks. Villanueva's model has been revised twice and each revision has narrowed the projected intersection window rather than widened it. Current best estimate: 20 June, ±1 day. The solstice is 21 June. I don't think the margin of error is real. I think it's the solstice.

Which raises the question we've been circling for a fortnight: do we intervene before the intersection, or do we let it arrive and see what happens.

The simplest intervention would be to remove the subject from the property before the convergence date. Fabricate a reason — family emergency, a work trip the client arranges — and get him somewhere else for a few days. If whatever is building requires his presence at that specific location on that specific date, disruption may be sufficient. This has precedent. It is the least invasive option and the one I would normally recommend.

The problem is that we don't know what we're disrupting. If this is a process — something with a beginning and an end — then interruption might resolve it cleanly. But if it's a summons, removing the subject may simply mean it finds another way. And we will have spent our best information advantage — the fact that the subject doesn't know we're watching — on a gamble.

I'm asking for positions before Friday.

Covert excavation commenced 18 April. Approach is by tunnel originating from the OP-1 scrub position and proceeding northwest toward the recorded centre of the 1971 dig site. Tunnel is narrow-bore — approximately 90cm diameter, timber-shored — and progress has been slow due to clay density and the need to operate without surface-visible spoil. All displaced material is being retained inside the monitoring post compound under tarpaulin.

On 21 April, at a depth of 1.4 metres and a distance of approximately 28 metres from the tunnel mouth, the team encountered worked stone. Three artefacts have been identified in situ so far: two flat, roughly rectangular slabs approximately 40cm in length, and one smaller piece, irregular, possibly a fragment of something larger. All three are sandstone, consistent with materials described in the 1971 excavation report.

The two larger slabs bear surface markings. These have been photographed and sent for preliminary comparison.

Regional seismicity for the area over the past decade shows three recorded events, all below M 1.5. A 2.0 is not unprecedented for southern England but it is uncommon, and the timing — within seconds of physical contact with the artefact — was noted by everyone present.

Hargreaves has requested a twenty-four-hour hold before any further extraction attempts.

I would like to formally note that I am no longer comfortable characterising these events as coincidental.

If the artefacts cannot come to us, we go to the artefacts. The terminal section of the excavation tunnel is being converted into a field study space. I am using the word "space" advisedly. It is not a laboratory. It is a widened section of a muddy hole in the ground, roughly 2.2 metres wide and 1.5 metres from floor to ceiling, thirty metres from the surface and accessible only by crawling.

Power has been run from the monitoring post via armoured cable. Lighting is a single LED strip and two battery-powered work lamps. Villanueva has installed a dehumidifier which is losing its war with the clay walls by a significant margin. The floor has been lined with plywood sheeting which buckles when you step on it. There is a folding table that cannot be fully unfolded due to the width of the tunnel. Chen has affixed a printed sign to the tunnel entrance reading "LAB" in what I assume is an attempt at morale.

Despite the conditions, we now have close-range photographic documentation of Artefacts A, B, and C, and work has begun on the surface markings.

The markings on A and B place this site within a known tradition of neolithic solar-ritual art, but with features that don't map neatly onto any catalogued site. We have something that resembles what has been found elsewhere without being what has been found elsewhere. Whether that distinction matters is above my archaeological competence. I have consulted with the Institution's archive section and they have recommended an external specialist.

I've been running the Artefact B markings against every published neolithic notation system I can access. The short-line groups don't correspond to any known lunar or solar calendar. They don't correspond to seasonal tallies. They don't correspond to anything in the Megalithic Art Survey.

What they do correspond to — and I want to be cautious here because I've been staring at these photographs for nine days in what is functionally a cave — is the photoperiod profile for this specific latitude. Not the solstices and equinoxes. The daily daylight duration, notch by notch, across what appears to be a full year. If I'm right, someone carved the length of every day into a stone five thousand years ago, and the grouping boundaries align with the dates on which the rate of change itself changes. Inflection points. Whoever made this understood not just that days get longer and shorter, but the mathematics of how they get longer and shorter.

I don't know what to do with this. Forwarding to Hargreaves for the specialist consultation.

The Institution has recommended a freelance consultant for the artefact analysis. His name — or at least the name he operates under — is Mr. Shine. He is expected on site 22 May. Some notes in advance, because there are going to be questions and I would rather address them now than in the tunnel.

Mr. Shine does not speak English. He speaks what has been described to me as an extinct South American language. I have been unable to identify which one. The Institution's briefing materials simply call it "his language" without further elaboration, which is a level of deliberate imprecision I have learned not to press them on.

He travels with a translator, a woman identified only as Ms. Caro, who interprets for him in all professional interactions. I am told the arrangement is long-standing and that Ms. Caro's interpretations should be treated as authoritative.

Mr. Shine will arrive wearing full-body covering. I am told this is not religious dress, not a cultural observance, and not a medical garment. It is described in the Institution's file as "professional attire appropriate to his specialism." He will also wear sunglasses at all times, including underground. Do not comment on this. Do not ask about it. Do not offer to adjust the lighting on his account. These instructions come from the Institution directly and I am passing them along without editorialising, although I will note that the file says "sunglasses" where I suspect it means something closer to "protective eyewear for our benefit."

I have worked with specialists referred by the Institution on two previous occasions. On both occasions the specialist resolved the identification problem efficiently and without incident. On both occasions I understood less about what was happening after the specialist's involvement than before it. I expect this will follow a similar pattern.

Mr. Shine's fee is being handled through the Institution. Do not discuss compensation with him or with Ms. Caro.

Mr. Shine arrived at the monitoring post at 08:40, accompanied by Ms. Caro. He was transported from London in an unmarked vehicle arranged by the Institution.

First impressions, for the record. The covering is not what I expected. It is not a robe, not a hazmat suit, not bandages — it is a tailored garment, modern in cut, covering him from the crown of his head to his feet. It fits well. It looks expensive. The fabric is dark, matte, opaque, and I could not identify the material. Over or beneath this — it is difficult to tell — he wears dark glasses. Not sunglasses in the casual sense. They wrap closely around the face, and the lenses are very dark. He wore them walking across the field at 08:40 in overcast conditions. He wore them entering the monitoring post. He wore them entering the tunnel. He is, I am told, wearing them now, thirty metres underground in a room lit by a single LED strip.

Ms. Caro is businesslike, mid-forties, and speaks English with what might be a Chilean accent, though I am not confident. She carries a leather satchel and a notebook and no other equipment. She introduced Mr. Shine by name only, declined tea, and asked to see the artefacts immediately.

Mr. Shine spoke to Ms. Caro at length as they approached the tunnel entrance. The language is unlike anything I've heard. It does not sound like Spanish, or Portuguese, or Quechua, or any of the other South American languages I looked up in preparation. It has a cadence that is difficult to describe. Osei, who has some background in linguistics, later said it sounded "pre-phonemic," which is a phrase I have written down without understanding.

They have been in the tunnel for four hours. Ms. Caro has emerged twice to ask for water. Mr. Shine has not emerged.

Mr. Shine exited the tunnel at 17:15. His covering was immaculate. I do not know how this is possible. Every member of the team emerges from that tunnel caked in clay from the elbows down. The plywood floor alone guarantees muddy knees. Mr. Shine looked as though he had spent the day in an office.

He spoke to Ms. Caro for approximately two minutes. She then turned to me and delivered the following, which I have reconstructed from notes taken immediately afterward:

"Mr. Shine says the markings on the first stone are a greeting. The markings on the second stone are a calendar, which your colleague has already partly understood. He says the third stone is not a fragment. It is complete. It is an eye. He says the eye is open."

I asked Ms. Caro what "open" meant in this context. She relayed the question. Mr. Shine responded briefly.

"He says: it is looking at the house."

Mr. Shine then walked to the Institution's vehicle and departed. Ms. Caro informed me that he would return tomorrow and that he had asked for the LED strip to be removed from the tunnel before his next visit. When I asked why, she said Mr. Shine preferred to work without artificial light.

I did not ask how he intended to examine five-thousand-year-old stone carvings in total darkness, because I felt that the question would not produce a useful answer.