KEG: Post II:
As noted in my prior post, Mortimer had a clear win after 25...Qh5. Amazingly enough, and even though Mason committed three serious blunders after 25...Qh5, Mortimer still managed to lose. Here's what happened.
This move, which allows the Black Queen a home at h4, was sheer suicide. 26. h4 was essential (though Mason would still have been lost).
26...Qh4! was crushing. It is hard to fathom why Mortimer issued this invitation for Mason to gobble up his strong e-pawn.
Mason had to play 28. RxR first. Leaving himself a Rook on c1 for Mortimer to exploit should have been a recipe for disaster.
29. Qg2 was essential. Now, as psmith has shown on this site (but as Rosenthal overlooked in the Tournament Book), Mortimer has a easy win with 29...Re8 (pinning the Queen). Apparently the players and Rosenthal missed what psmith noticed: that the pin remains effective even after 30. RxR since Mortimer could then have played 30...Qf3+. In short, Mason's Queen is hanging here.
Bravo psmith for spotting this!
As noted above, Mortimer had an immediate win of the Queen and of the game here with 29...Re8! But, even after this blunder, Mortimer still had a won game.
Only now is Mortimer's win gone. psmith has correctly noted that Mortimer should have played 30...Qh5. But I disagree with his overall assessment of the position after 30...Qh5. Black still would have had a win had he played psmith's excellent suggested move. After 30...Qh5 31. Rh1 (the reply given by both Rosenthal in the Tournament Book and by psmith in his otherwise fabulous comments on this game), the killing response by Black would be 31...Bh2! rather than 31...Re8, though even that should yield Black a win. After 30...Qh5, 31. Qe6+ would be best, but I still do not see how Mason would have saved the game.
In any case, I certainly agree that 30...g3? was a blunder after which Mortimer no longer had any advantage. But he certainly should not have lost this position.
The beginning of further troubles for Mortimer. 31...Qc8 was best.
Mason throws away most of his advantage after this move. 34. fxg3 was much better.
After this, Mortimer is lost. My guess is that Mortimer saw ghosts here. The simple 37...Bxg3 yields a nearly equal game. Mortimer probably feared 38. Rg1, but then 38...Rh8! saves the day for Black.
A pretty move by Mason which wins the game. The idea of setting up a skewer on the a3-f8 diagonal is fascinating. Mortimer did not see what Mason was doing until it was too late.
As psmith correctly states, Mortimer's only chance here was 38...Kf5 (much better than 39...Kf7 as suggested by Rosenthal in the Tournament Book).
My guess is only now did Mortimer notice the skewer combination Mason had hatched up. Mortimer could have resigned at this point.
An interesting sidelight is that Mason still had a chance to go wrong even after 39. Rd1+. Had Mortimer played 39...Kc5 the tempting 40. RxB loses to the intermediate move 40...Re8+. Presumably, however, Mason would have played (if 39. Kc5) 40. Ba3+ after which it is game over.
Now there is no way for Mason to go wrong.
Mortimer decided to play on though down a piece in the endgame!
Mortimer did manage to win a piece after Mason allowed him to set up a pin on the g-file, but after 48...RxB 49. RxR+ KxR, the King-and-pawn endgame was an obvious win for Mason, and Mortimer finally gave up on move 51.