Hastings, 23 August 1895. Marco's Ruy Lopez yields Black an early open game, and Dr. Tarrasch, after a temporary sacrifice, conducts a model king's-side attack that produces an announced mate. Notes by S. Tinsley.

1. 6.Bxc6+ Bearing in mind one essential principle of all close games—viz. the importance of keeping the opposing forces shut up—it can scarcely be advisable thus early to exchange, opening at once a file for the Queen's side Rook, &c. The usual P to Q 4 is obviously superior at this point. Or White may even play P to Q 3, followed perhaps by P to Q B 3 (to preserve the King's Bishop). Q Kt to Q 2, Q to K 2, and R to K sq are also occasionally adopted here.
2. 7…Nd7 Dr. Tarrasch's patent. It looks awkward, and so it is.
3. 8.dxe5 Still further opening up the game for Black. Kt to B 3 is clearly the correct course, and would develop White's game, leaving Black to get his Pieces free as best he can.
4. 12…Nb6 R × P would lose the exchange, as White replies Kt to Kt 3 at once.
5. 15.b4 This move is deceptive. At first sight good, because, of course, the Pawn cannot be taken safely; but on examination the effect will be found to be to enable Black to dispose of his doubled Pawn and gain still further liberty, leaving the Queen's side Pawns (White) isolated. He might have aimed at getting a Kt to Q 5 by (1) Q R to Q sq; (2) Kt to Q Kt sq; (3) Kt to B 3, and afterwards to K 2 or Q 5, as needed by the circumstances of the time.
6. 16…O-O 16. ... P × P; 17. P × P, R × P, would give White a strong passed Pawn on the Queen's Rook's file.
7. 18…f5 Students of Dr. Tarrasch's play will be struck with the effective simplicity of his moves when the critical time arrives. So here. The whole scene changes in a moment, and the game is all in Black's hands. Every Piece, too, is in its right place. The position is worthy of special note. (See diagram.)
8. 22.Nxe4 22. Q × B, R × Kt (!); 23. Q × Q, B × Q; 24. B × R, B × B, and Black has gained two Pieces for the Rook.
9. 22…Bxh2+ The sacrifice appears to be of questionable merit, though it yields at a critical period an immediate attack.
10. 25.f3 We suggest for White now P to K B 4 instead, and we imagine White has a valid defence against all immediate attacks. The variations are somewhat obvious. Here is one: 25. P to B 4, Kt to K 4; 26. P × Kt, Q to R 7, ch; 27. K to B 2, R to K B sq, ch; 28. Kt to B 6, ch, &c.
11. 26.Bg5 Apparently to prevent the attack by 26. ... Q to R 7, ch; 27. K to B 2, R × P, ch; 28. K × R, R to B 4, ch, &c.
12. 31…Nxf3 This remarkably fine move soon leads to a final breakdown of the defences. It is safe to say that few besides Dr. Tarrasch would have discovered this way to a conclusion.
13. 41.Kf1 It is fair to point out (I was looking on at the time) that Herr Marco was inclined to resign at this period, but in the general interests of competitors played to the end like a man. A curious game, very interesting. White, in spite of a poor opening, might, I think, have done better. The game produces a not easily defined impression that there was 'a way out of it' after the sacrifice at move 22.