Thanks so much, Robert, very helpful.
I attach some more pictures. I'm a lousy photographer with a lousy camera, so can't do really good close-ups. (Also: the yellow looks too electric on the photos, the other colours have the roughly right shade but are a little too bright.)
Two things I couldn't get on the photos are these:
(i) it's almost as if there are two layers of warps, one on top of the other. I mean, at the ends, it's as if almost exactly under every warp there is another warp.
(ii) the pile (very short clipped) has almost no direction, even where it's full, rather almost stands up straight. Not completely of course, you can still only stroke it in one direction.
On age of the rug:
(i) the rug is not top-bottom symmetric, in that the medaillon pendants are one heart shaped, one drop shaped. I've only ever seen this on pre-1910 rugs. It seems after that, at least for commercially produced rugs, if there is a centre medaillon, left-right and top-bottom symmtry is a must. (As you have it when you use a quarter cartoon). Some rug books say this kind of pendant asymmetry (which is common on safavid rugs) was meant to symbolize sun and moon. I'm not sure I believe this, but in any case, the asymmetry is not accidental.
(ii) the main-border corner solution: In pre-1880 Tabriz rugs there is often no corner solution at all, just as in tribal rugs. Roughly after that, weavers seem to try and put a diagonal version of amain border motive in the four courners, but in pre-1900 rugs as a result often have to compromise the vertical main-border design towards the end. On commercially produced rugs round 1920, that never seems to happen, rather there are perfect corner solutions (as you'd get them with quarter cartoons).
So I was putting the rug as pre-1910, but I'm happy to be proven wrong, I like it anyhow.
Shereen attached the following image(s):