I'm always in the market for those ambitious repertoire books that offer a complete system of play for one side or the other. I want to see what the authors, who are usually strong players, are recommending against the systems that I play, and in some cases, I find good ideas that I can add to my own repertoire. These books are especially good for recommendations against silly systems like 1...b5. But it's surprising how often the authors recommend suboptimal play. While several authors recommend the Italian game, I don't know of any that continues with 4. Ng5! against the Two Knights Defense. Yet 4. Ng5! is by far the most serious challenge to Black's system.

The system presented in Chris Baker's A Startling Chess Opening Repertoire, Cadogan 1998, is a case in point. The author, an English IM, recommends 1. e4 e5 2. Nf3 Nc6 3. Bc4. His idea after 3...Nf6 is to steer towards a Max Lange Attack with 4. d4 exd4 5. O-O. Since I have recommended 1...e5 in this space, and since I'm particularly keen on the Two Knights, I would like to examine Baker's ideas here.

From the diagrammed position, I have generally played 5...Bc5, acquiescing to the Max Lange, because that famously difficult line gives Black real winning chances. But while 5...Bc5 may be Black's best winning try, it is both dangerous and torturously complicated. But most solid and also completely adequate is 5...Nxe4! which, for some reason, is being called the "anti-Max Lange" by Baker and some other authors. That is a little like calling 1. d4 Nf6 the "anti-Queen's Gambit." Anyhow, 5...Nxe4! 6. Re1 d5 7. Bxd5 Qxd5 8. Nc3 and now I believe Black's simplest and therefore best path to equality is 8...Qh5! (although 8...Qa5 gives White the complications he is seeking, it is also sufficient for equality) 9. Nxe4 Be6 10. Bg5 (10. Neg5? is a common error after which 10...O-O-O 11. Nxe6 fxe6 12. Rxe6 Bd6 leaves Black with the upper hand) 10...Bd6! and further:

A. 11. Bf6 O-O 12. Nxd6 cxd6 13. Bxd4 Bg4 and Black has equalized, for example, 14. Re3 Rae8 or 14. Bc3 d5.

B. 11. Nxd6 cxd6 12. Bf4 Qd5 (12...Qc5 is also playable) and now:

B1. 13. c4?! Qxc4 14. Bxd6 Qd5 and Black will castle to whichever side White's bishop retreats, with an excellent game in either case.

B2. 13. c3 Rc8 14. Nxd4 Nxd4 15. Qxd4 Qxd4 16. cxd4 Kd7 and Black was fine in Wikner-Wedberg, Osterskars 1995.

C. 11. c4 is the move recommended by Baker, based upon Sveshnikov-Bezgodov, St. Petersburg 1994, which ran 11...O-O 12. c5 Be5! 13. Nxe5 Qxd1 14. Raxd1 Nxe5 15. Rxd4 f6 (15...Bxa2? 16. Be7! Re8 16. Nf6!! gxf6 17. Rxe5!) 16. Bf4 Bxa2 17. Nc3 Bf7 18. Bxe5 fxe5 and now, instead of 19. Rxe5+ as played in the game, with equality (the game continued 19...Rae8 20. f4 Rxe5 21. fxe5 Be6 22. Nb5 Rf5 23. Re4 Rf7 24. Nxa2 c6 25. b4 Rf8 26. b5 cxb5 27. Nxb5 Rc8 28. Nd4 Kf7 and Black had no trouble holding the draw), Baker says 19. Rd7! "would have kept an edge." But that is a point that I would like to debate. First, 19...Rac8 is obviously called for.

Analysis position after 19...Rac8

Now it appears to me that Black, with care, can expel White's rook, after which Black actually has a little play for a win based upon his better minor piece, the possible weakness of White's overextended c-pawn, and prospects of creating an outside passed pawn. For example:

C1. 20. Nb5? Be8

C2. 20. Red1? Be6 21. Re7 Rfe8 22. Rxe8+ Rxe8 and the a-pawn can't be won without trapping the knight: 23. Nb5 c6

C3. 20. Nd5? Be6 21. Ne7+ Kf7 22. Nxc8+ Bxd7 23. Nxa7 Ra8 24. Ra1 c6 and the knight is trapped (25. Ra5 Rd8).

C4. 20. Rxe5 Rfe8! (20...Rfd8? 21. Rxd8+! Rxd8 22. Re7 Re8 23. Rxe8+ Bxe8 23. Nd5 c6 24. Ne7+ Kf8 25. Nc8 and White is better) and now:

C4a. 21. Rxe8+ Bxe8 22. Re7 Kf8 with at least equality.

C4b. 21. f4 Rxe5 22. fxe5 Be6! 23. Rd4 (23. Re7? Kf8) 23...Kf7 and Black's game looks fine.

C4c. 21. R5e7 Rxe7 22. Rxe7 Kf8 23. Rd7 Be6 and Black is equal or better.

C5. 20. Re7 Rfe8 21. R7xe5 (or 21. R1xe5 Kf8 22. Rxe8+ Rxe8 23. Rxe8+ Bxe8) 21...Rxe5 22. Rxe5 Kf8 23. f4 Rd8 and Black is at least equal.

I would very much like to hear from anyone who has anything to contribute to the evaluation of the position in the foregoing diagram. The position after Black's fourth move is important for the theory of the open games outside the Two Knights, since it can also arise from the Scotch Gambit (1. e4 e5 2. Nf3 Nc6 3. d4 exd4 4. Bc4 Nf6) and Bishop's Opening (1. e4 e5 2. Bc4 Nf6 3. d4 exd4 4. Nf3 Nc6).

But as for now, I maintain my conviction that 4. Ng5! is White's best try against the Two Knights. Oh yes, it is entertaining for Black to play against, especially if he is as enthusiastic a follower of the great Tarrasch as I am. But, notwithstanding the great Doctor's perfectly correct observation that "this move surrenders the initiative, the birthright of the white pieces," White can play 4. Ng5! with good hope of winning. There is a fine balance between White's extra material and Black's superior piece activity in these lines, making them a good vehicle for either player's winning aspirations.

As to how Black should answer 4. Ng5!, the clear favorite is the main line, 4...d5 5. exd5 Na5, though not everybody agrees. Back in 1969-70, former World CC Champion Hans Berliner wrote a series of columns in the Washington Post in which he espoused "The System," an approach to the opening that placed a great deal of emphasis on the fight for the center and the vigorous use of the pieces. Concerning 5...Na5 Berliner wrote, "this move decentralizes a well placed piece in favor of a momentary attack, and thus is not at all in accord with System principles." He continued: "It is more appropriate instead to think of mobilizing new forces while counterattacking in order to take advantage of White's loss of time." And thus he advocated the Fritz Variation, 5...Nd4, a line which he continuously championed in his practice, with good success.

I won't try to plumb the theoretical depths of the Fritz Variation this month, but for what it is worth, the judgement of modern theory is that 5...Nd4 confers significantly more winning chances upon White than 5...Na5 does. The future will decide who has erred in judging the Fritz Variation: Berliner, or the chess world. But until somebody can show me how White can get the advantage against it, I'll continue to play 5...Na5!, not particularly happy to place my knight on the rim, but very happy to have the initiative, a significant lead in development, and open lines for my rooks and bishops. In my chess practice, I've found these assets to be sufficient compensation for the pawn and the inefficiently positioned knight and the queenside pawn weaknesses that Black accepts in this variation.

After 5...Na5 6. Bb5+ c6 7. dxc6 bxc6, White has two playable moves, 8. Be2 and 8. Qf3, respectively illustrated in the two games given below.

Game 1. Rohde - Morss, WTGM365.

Game 2. Castaldo - Morss, USCF-92CM96.

Next month: the Modern Defense to the King's Gambit.