Game 46 in Philip W. Sergeant’s Morphy’s Games of Chess (1916), printed pages 98–100. Commentary from this edition; named earlier annotators remain credited in the notes. Sixth game of the 1858 Harrwitz match. The source says the time was not recorded.

1. 4…Nf6 Inferior to both Kt—QB3 and B—Q2. Nevertheless it is not hopelessly bad, especially when White continues as in the present game.
2. 5.e5 Steinitz does not approve of this, and gives 5 Kt—B3, Kt—B3; 6 B—QKt5, B—Q2; 7 B×Kt, B×B; 8 B—K3, B—K2; 9 Castles, Castles; 10 Q—B4, Q—B1 as the best play on both sides (M.C.I., 148).
3. 5…dxe5 Here is Harrwitz’s real mistake. 5..Q—K2, and if 6 B—K3, Kt—Kt5; 7 P×P, Q×P was correct (M.C.I., 149).
4. 8…Bd6 Why not B—QKt5? It would obviate the evil choice which Black has next move of developing White or getting an isolated Pawn on the Q file.
5. 11…Ke7 Kt—Kt5 is better.
6. 19…Rb8 R—R2 is even less promising.
7. 24.Kb3 P—QKt4 was better, because, as the game went, Black could have checked with Knight on move 25, driving White’s King to R3. Morphy lost time over the King’s excursion.
8. 33.Bxe6 Black throws away excellent drawing chances by this, while White now handles the end-game with great skill.
9. 48.Qf3 If 48..P—R7, 49 Q—Kt3ch, K—R8; 50 Q—B2, mating next move.