I didn't see it, but Mick Pork has mentioned on the previous thread that Andrew Neil has been abusing his position as a BBC host to crow about a poll that purports to show that Scots want Trident to stay here. As Neil always takes such sadistic pleasure in taking apart interviewees who don't have a proper understanding of the numbers (and somehow got away with it in a recent interview with Douglas Alexander even though he'd made a catastrophic schoolboy slip-up in his own sums), I wouldn't mind seeing someone do the same to him in respect of this poll. This is what he appears not to have taken account of -
1) It actually shows that the Scottish public do not want Britain to have nuclear weapons at all, by a margin of 46% to 37%. (That makes it uncannily similar to an Ashcroft poll which was also brazenly misreported as showing that Scots were pro-Trident.) It's impossible to argue that the public want Trident to stay when they do not think it should even exist.
2) The question that has generated such excitement among the nuclear weapon loving Brit Nats (Johann Lamont must be so proud) implicitly presupposes that Westminster will be retaining the inhuman weapons against our wishes, and merely poses a practical query about what should happen to them after independence.
3) This is not a full-scale Scottish poll at all, but a small subsample of a Britain-wide poll. It's therefore of limited validity. Even if it had been a full-scale poll, the tiny 4% lead for keeping Trident on the Clyde after independence would have been within the standard margin of error. As it is, the margin of error for a small subsample is incalculable, and it's therefore entirely conceivable that the same question posed to a full-scale, properly weighted Scottish sample would produce the opposite result.
Incidentally, the Herald's reporting of it as a "new poll" is also somewhat misleading - it may have only just been released, but the fieldwork for it concluded last October!