Special, Like Saturday Night!

I do not know that stretch of line too well.
London Clay is quite a slippery substance, as is fresh grout, and I think that these factors contributed to the initial movement that I described. I have never really thought about the possibility of a tunnel twisting after construction, but it does seem theoretically possible. Does anyone else have any ideas or knowledge of this?

I agree completely that settlement will be minimised. What we were wondering is whether a tunnel might rotate slightly within its bore through the clay after construction had been completed. This is something that I had never considered.
If the track was banked slightly, there would be greater weight on one side of it. Would the vibration caused by the movement of the train, coupled with this extra weight allow any rotation, however small?


And when it dries, it gets very hard and brittle.

There were sidings to the south of Stockwell station by this date, but they were never used in this way. Incidentally, there was also a little known running tunnel that ran parallel to the station at Stockwell and allowed a locomotive to run round from the back to the front of a train if required.
There was also a siding between the running tunnels at Elephant & Castle.
This photo appears not to be a siding. I am sure that this is a running tunnel. But where?
Expect the unexpected.
May I suggest another inspection of the area in the photograph in which a signal has been identified. This is worth a closer look.

Time to reveal all.
All of the evidence – including the 1899 magazine article that this picture came from – suggested that this was a picture taken in the southbound (down) tunnel outside of King William Street (KWS) station. For many years, that is what I thought until I analysed the evidence.
I assumed that what I could see was a train that had left KWS, photographed from the locomotive that had originally pulled it northbound and had followed it a short distance out of the station before returning to couple-up to the next train that was reversing there. This would make the signal that we can see the back of a shunt signal to call the loco back into the station (note the small ventilation slot that indicates it is the rear of a Dutton tunnel signal).
The thing that always worried me was that the southbound tunnel leaving KWS swings immediately into the sharp Arthur Street curve. The first stretch of straight tunnel is two train lengths away.
I know that locomotives performing a shunt entered the tunnel just enough to clear the crossover and then stopped, so why was the signal such a huge distance from the station? I eventually located a signal diagram for this section of track that showed no shunt signal at the south end of Arthur Street curve. This then was not that location.
After checking the entire southbound line from KWS to Stockwell, I found nowhere that fitted this view. I then considered the possibility that this might have been shot on the northbound line with a locomotive that had been uncoupled from the train. This seemed an odd thing to do until I spotted something just above the signal and realised that the location appeared to fit what I was looking at.
tubeoperator92 has previously mentioned how trains that were unable to get up the steep incline into KWS were often assisted by a second locomotive. We must remember that in the routine operation of the train service, a locomotive would pull a northbound train into KWS station, uncouple from the north end, allow the train to be reformed for its journey south, follow the train as it left the station, stop clear of the pointwork on the southbound track just outside of the station, wait for the arrival of the next incoming northbound train, run in and couple up to the south end of this train to form the next southbound train. By this time, the incoming locomotive would have uncoupled from the north end of the train and the process would be repeated. This meant that, although there was no space at KWS for a spare locomotive, there was always a loco somewhere in the vicinity of that station that was not coupled to a train. In the early days it was this loco that was sent back from KWS to assist any train that was in difficulty on the approach. This though was problematic, as there was barely enough space in the platform for two locomotives, three carriages and a fresh locomotive to pull the train south. The other problem was that if the assisting loco had pulled a train all of the way into the station, it would no longer be available to head the train out as it would be trapped in the platform.
As a result, a method was worked out to call-on the loco from the following northbound train to use as a banking locomotive, to help push the train into KWS. After it had assisted in this way, it returned south down the northbound line to retrieve its carriages and passengers!
In cases like this, the loco that was called-on to assist was whichever one was held with its train at KWS Outer Home signal, (this signal was a long way back on the other side of the river just south of the SE Railway viaduct at London Bridge). In order to tell the locomotive that it was needed for this unsignalled, unofficial movement, the company installed a telephone at the Outer Home and phoned the driver.
The location that you are looking at is (I think) the KWS Outer Home signal. The telephone can be seen just above the signal with the upper of the two cables running from KWS and looping into it. The bottom cable is a telephone line to Borough northbound signal cabin.
The large board fixed to the tunnel wall on the left of the picture has a frame around it (I have a larger and slightly higher definition picture than the one posted) and you can just see the light reflecting off the bottom edge of this frame. There appears to be a sign on this board, but it is illegible in the photograph. It is placed at cab height and I suspect it to be an instruction to the driver – maybe a warning to secure the carriages before proceeding forward as a banking engine - who knows?
This is of course a logical place to uncouple a locomotive for a photograph. The drivers would have previous experiece of uncoupling at this location and there was even a telephone to contact the signalman. Far better to take a picture from the safety of a locomotive than risk sending a photographer with a huge glass-plate camera, flash-tray and tripod into a damp tunnel next to a live rail.

Thank you, CSLR, for posting another insight into the operations of this fascinating railway company.

Just to clarify. There are two points being discussed in this thread both of which stem from the same photograph. The other one relates to the location, while this one relates to the asymmetry of the segments.
I am convinced that the misalignment occurred during construction and do not think that anything significant occurred subsequently. The comments of stanmore K appear to reinforce this view.
I am currently searching through my large collection of photographs as I am sure that I have an unprinted negative somewhere that highlights part of this point. If I remember correctly, this shows a point in the tunnel where the construction engineers 'side-stepped' a ring of tunnel segments in an attempt to move everything back into the correct position.

Just to tidy up a few points that were mentioned in this thread.
The original tunnels on the C&SLR between King William Street and Elephant & Castle were 10' 2" in diameter. From Elephant & Castle to Stockwell the tunnels were 10' 6".
Regarding the comments made about the signal. The original tunnel signals were made by Dutton & Co and had a single lens. The aspect was changed by rotating a frame containing coloured glass horizontally around the body. This is the type of signal seen in the photograph. As the line was extended, the new signals had aspect lenses that moved vertically and were of a similar design to those used on the LER system.
Finally a note regarding terminology. Although I have referred to the lines by the modern convention as north and south, the company always called them Up (northbound) and Down (southbound) at this time. Similarly, the C&SLR always operated 'carriages' on its trains. 'Cars' was a later term that came in with the other tube railways.