Pat — OK, just for you.
Back in 1998 Sotheby's made the same assumption as Adam above — that the Royal Brierley factory collection contained Royal Brierley / Stevens and Williams glass. However the Waterlily epergne carried a design registration number 409769 punched into the gilt metal stems. This registration was made on 9th May 1903 by John Walsh Walsh, not Stevens and Williams. Incidentally, after the auction Simon Cottle, head of glass at Sotheby's, publicly apologised for the error, a record in itself, as I have yet to hear of any other occasion where an auctioneer apologised for getting it wrong. This makes me his number one fan! See Reynolds pl. 10 for a small version of this epergne.
Virtually all English late Victorian or early C20 Stourbridge-type unattributed fancy glass is described by US regional auction houses as either Webb or Stevens and Williams, whereas here in Britain our auctioneers choose from a longer list, which has the strange effect of making their attributions seem more authentic. The Dallas epergne was again by John Walsh Walsh, in a type of glass Walsh called Sateen, launched in the June 1886 edition of the Pottery Gazette. Eric Reynolds discovered positive proof of all this in 2000 in the records of a Sheffield electroplating company, since when Walsh Sateen glass has become reasonably well-known and collectable in its own right.
So that's it. I look forward to the day when Walsh auction records are set by Walsh glass actually attributed by the auctioneer to Walsh, but that could still be a long way off!
My compliments to you all on your ingenious solutions.
Bernard C.