blackwolf wrote:

It is quite possible to have power to the brake lights but no power to the tail lights, but it does tend to suggest that there is an earth connection albeit not necessarily the correct earth connection!

Certainly is, which likely denotes that the ground is probably okay, but could be due to varied impedance between the two circuits.
This is why I’m wondering it’s possible, that a female conned for on one of the male pins on the main beam stalk might have popped off.
Reason being, is the sidelight and dipped switch runs its power up through there, as well as main beam.
Main beam and brake light are separate, and working but, if one of the other connections are missing that could be the missing link in the circuit and would also mean the fuse would be in-tact as it wouldn’t blow as the pins are static and the female socket insulated so it’s unlikely there would be a short circuit to blow the fuse but the circuit broken all the same.
The only way of checking that with a multimeter would be impedance check from the ATO fuse, to the tail light pin, if there’s high impedance or no reading then there is a broken circuit there.

And it would be a single connection for all the lights, I doubt it’s a broken wire or split insulation and cracked conductor as that would be only on one side.
I do know on a Puma at least that the exterior lighting ground is on the side the transfer box, and is for front and rear, with two ground wires and two ground eyes, presumably one for each side.
As far as I’m aware, they are the ground for the indicators, sidelights and tail lights and likely the brake lights also.
And that particular flanged I think m6 bolt, liked to come loose by its self, probably due to being removed with a clutch change and vibrate loose it’s self and cause a poor ground on exterior light circuits, and cause all sorts of strange issues such as if you indicated (flashers turned on) the opposite side would light up too.
The particular bolt in question would be torqued to the WSM spec, but still come loose.
So in the end I put a little Loctite 222 (low break off strength thread locker on the threads) and it’s stayed put since, I like that loctite thread locker but I’d not recommend any others due to being too strong. It’s ideal just to ensure something doesn’t come loose in the absence of not being able to practically use a spring washer or nyloc nut or similar locking fastener.
In the WSM, I seem to recall those two ground points being called the ground for the rear lighting loom, but a tall also was for front too, one ground for each side on each grounding eye on the one bolt/stud.

It was the source of quote a headache issue to me once, that reoccurred twice until I tracked it down.
And I tracked it down by doing an open impedance check on each indicator ground and then the reading was high, but wasn’t to a separate god ground I knew it was common ground issue for both sides so led me to a common ground as above.
Normally live issues are a popped off connector, or a short and the latter causing a fuse to be blown so I don’t think a short is the issue here.

One issue can be is that Multimeter test leads aren’t always long enough to test from A to B, but there are some ways around that with test leads.
Little alligator clip test leads can be useful, but also not even long enough sometimes but you do have to take into account the losses from having them in the circuit.
Impedance checks I find are more useful in the unknown than voltage, but best done with power completely isolated and turned off.
I’m thinking loose live or popped off connector in the dipped, sidelight switch area or indicator stalk mech,* especially the latter here.*

It is established brake lights on that circuit is fine, indicators are fine, so grounding is there for that so suggests it’s likely that is okay.
But a common live is missing, a single common live and if that was present and fuse isn’t blown, then a connection maybe missing, either * or behind the fuse panel.

Hope the OP manages to find it, they are difficult to find for sure.
If the original lights are refitted and still don’t work, then it’s probably not impedance but a connection disturbed and misusing causing a broken common live circuit.

I’ll need to rest my head now.  $W33T $0U7H3RN $UG4R
🇬🇧🏴󠁧󠁢󠁥󠁮󠁧󠁿🏴󠁧󠁢󠁷󠁬󠁳󠁿🏴󠁧󠁢󠁳󠁣󠁴󠁿🇮🇪🇺🇸⛽️🛢️⚙️🧰💪