As I sat and depressed the clutch pedal from the driver's position, I noticed that the whole of the engine and gearbox was lifting and tilting on its rubber mountings. Not good; back to the drawing board.

Once again, I removed the gearbox, bellhousing and pressure plate to see if there were any clues I'd missed the last time as to the heaviness of the clutch. The Ford plate on the left is smaller in diameter than the original Morris plate and the centres have different numbers of splines. The Morris centre is 5mm taller, but that affects nothing as the release bearing and its opposite number on the pressure plate don't even come close to touching it. Nevertheless, I sent both plates down to Precision Clutch Components Ltd to have the centres swapped and the Morris plate relined. Both plates are Borg & Beck, so it was a straightforward job. I'm happier now I can use the larger plate.

In the original clutch operation, the pedal is connected via two rods and a bellcrank to this lever which is located on the release bearing's shaft in the bellhousing.

My chain-operated fandango was necessary because the pedal's fulcrum in the pedal box was now at the top. I opted for this angle on the clutch lever thinking that it would be easier to lift from below the horizontal. An extension added to the top sprocket effectively doubled that diameter but made not the slightest difference in effort required to disengage the clutch.

Breaking out the disc cutter and welder, I tried again. No appreciable difference - still way too heavy. Holmes, who's been following my enthusiasm for the weirder solution, has pointed out that by the substitution of a rod from the top sprocket (do away with the chain) and a bellcrank mounted and connected to another rod so that the action is then horizontal to a lever vertically aligned with the release bearing, I may have more success and cure the heaviness too. I think Holmes has a point.

And on the subject of headaches, The Great Collector's Crossley simply refuses to start. Easy Start did not produce the goods. The tappet adjustment was proving particularly irksome as despite correct procedures, some of the gaps doubled after a couple of turns of the crankshaft. I was wondering if the cups in the cam followers in which the rods sit might be the cause, but as they're way down in the block and even to remove the rods means taking the head off again, I might leave that investigation as a final desperate act.

Inspecting the Stromberg carb seemed a good place to start (ha!) as the sparks department would light up Blackpool Tower. The idle jet was in two pieces, so I turned up a short, tapered sleeve to solder them together. The float was full of old fuel necessitating the delicate job of emptying and re-soldering. Those two problems might have some bearing on the engine's reluctance. After that, I've run out of straws.