I don’t know what this guy was looking at, but it wasn’t the shadow of the tip of the Shard. In the five days since the last similar shot, the gap had moved slightly south and west, according to the conveniently-placed rows of dots on the road.

This was the photo that closed the loop, the last shot taken before the Sun got back to where it was in the sky when I started.

I knew roughly where these shadows would fall, but it still amused me actually going to find them. To slot this final one into place, with the map revealing both the inevitability of the analemma and the shonkiness of my cartography, gave me a rare feeling of joy at a project seen through. A feeling that was clearly premature, as I let the blog sit idle once the city and the world reopened, and didn’t finish the photo blog, the map or the description.

The upcoming eclipse has got everyone here thinking about celestial mechanics, and you can’t say that about many things. It got me thinking about this again, and it’s a good time to close the loop.

In the months after taking this photo, I continued to head down to the shadow to fill in some of the yawning gaps in my analemma, closing it despite its inevitability. These photos remain to be added to the map and the blog. There’s another eclipse in Europe in August next year. Maybe that will motivate me.

All those shots of the Sun through the tip can get a bit samey, and it’s good to have road markings rather than river to anchor the shadow’s gap to the map.

By this time, nearly a year after starting this, the shadow was back on the same busy stretch of road where it started, with thousands of journeys detangling through the exit onto the street and the fast-passing shadow cast on it.

A scorching summer day, a deep blue sky and a sweaty cycle down. The pressure had been high and temperature had been building all week. No chance of a stray cloud today, although the view was nearly blocked by a hanging basket of flowers.

Nearly a month later.. Despite an unpromising layer of grey, the cloud was thin enough that the Sun was clearly visible through the gloom and the Shard tip. This time it was on the eastern side of St Olaf’s car park, a good way east of its July sightings last year.

I’d been a bit concerned that the narrowness of the small tip of the figure eight would mean there wasn’t really much visible difference from the ground on the scale of my map, but I was a good dozen or so metres from those first few measurements and the plot was starting to look like it might work. In mid-May the Sun is at its westernmost of the summer loop of the eight, so the difference between this bit of the loop and its easternmost side, in mid-July, is the greatest. It doesn’t translate to much on the ground, but enough.

You’ll note the Sun is not visible through the tip of the Shard, from the south side of St Olaf House. And as I suspected on my chance passing on 17th April, the shadow had long left the river. This was taken at 12:56:

No tip visible on the river, and no Sun visible through its prongs. The shadow must fall between these two points, on the roof of St Olaf’s House. So unseen I pinned it to the map there.

This stands as a bit of a bonus shot. With lockdown partially lifted, and the brightness less matched by the coldness, we’d arranged to meet friends miles away on Clapham Common, and we cycled. And we were late. I’d never cycled this route before, down the no-cycle-lane Borough High Street towards the turbulent and daunting Elephant & Castle roundabout, but our lateness made the timing for this shot perfect. Accidentally, I can assure you.

If the shadow tip was only just visible the day before, this was surely its last hurrah. And so it proved.

It was a bright cold day in April, and the clocks were striking 13. Well, 12:56 when I arrived, but 13 shortly afterwards. Annoyingly, despite it having been a cold clear night and yes, a bright cold morning, it clouded over as the day progressed. By the time I got into position the Sun was behind a large slow-moving cloud and there were no shadows anywhere, save the wide splotch of the cloud itself. It did gradually move though, southwards in the chilly northerly wind, and the shadows appeared 90 seconds late. Still, given the inexactness of my map marking when it’s on the river, that’s probably within the bounds of error.

It was only the very tips of the forked top that were visible now, dual dark spikes just peeking over the flat-topped shadow of the wharf.

It was a crazy week at work. But amidst the whirl of chaos, it was the first sunny midweek in a week, the forecast for the next few days was bleak, and I knew the crossover point was close.

The crossover in the figure 8 looks, from the plot, to be about 11/12^th April and 29^th August. Having failed miserably to get anything close to the point in the late summer, this one compelled me to leave my ailing database for 45 minutes and head to the bridge. I cycled again, stressed.

This time there wasn’t the luxury of a leisurely walk south, but I’d prefer to call it perfectly timed than a rush. As I arrived a tug dragged its huge load downriver, but the shadow was so close to the south bank that it didn’t obstruct the view. And it was a fair bit further south for its week. Attempting to triangulate again, it now aligned Belfast’s bridge with the Clipper pier, but it’s trickier the other way. The map marker is a best estimate.

The exact crossover point I feel should be celebrated. It’d be possible to work it out exactly. I wonder what it would take to get the Port of London Authority to put a buoy there, topped with a disco mirrorball, to reflect the moment to everyone around?

(Incidentally, although I’d written the above post several years ago, I’ve only picked this up again in August 2026. Why now? Maybe the looming eclipse puts such things in mind. I’ve also mentioned it to a few people now, and the fact that it’s unfinished irks me in an unusually tugging way.)