I've finished my testing and have spoken at some length with the importer, and have calmed down a bit.
The measured white point of the old and new batches seems fairly close - only delta E of around 0.2. Despite that there's definitely a small but clearly noticeable warm shift when printing colour, and also when printing monochrome using QTR-K3. Re-profiling deals with the colour printing case, and for K3, I assume that adding in a small amount of the cool curve would do likewise. Most of the shipped QTR K3 neutral curves are slightly warm rather than dead neutral anyway, IMHO.
For Piezography, I print using Special Edition, and whatever colour shift there is must be small compared to the SE effect. Actually if anything my SE test print was every so slightly less warm using the new CRP batch than the old, so go figure. When checking linearisation, the difference between the a,b numbers for the two batches was smaller on my Piezo printer than on my colour printer, which is what I'm seeing in my test prints.
There were also linearisation differences between my Piezo and K3 printers. My P2 curve for CRP is still linear on the new batch, but my K3 curve isn't. The new-batch K3 curve tracks much the same curve as the old batch down to the 90% patch, but then dives to a much darker dMax at 100%. Go figure.
So there's no sign of John Dean's mottling. The importer was adamant that despite shifting a huge amount of stock during this sale period, no-one else had mentioned this. He was also adamant that although some Canson production has shifted, it's only three papers and CRP is still being made the same way with the same coating. So all I can conclude is that what I'm see is batch variation. There's no sign at all of John Dean's mottling. The fact that he is getting on both CRP and the Epson Legacy version is puzzling.
Here endeth the report