R × P would make it practically impossible for White to win, if he can win at all. White's best continuation then would have been: 39 B - B 4, R - B 7; 40 R - Kt 5 ch, K - B 2; 41 B - Kt 8, P - R 6; 42 P - R 5, P - R 7; 43 B × P, R × B, and if there is a win it is very difficult to find it, as against 44 P - R 6, R - R 3 ! offers excellent chances for a draw.

If 40...P - R 7; 41 R - Kt 5 ch, K - R 3; 42 R - Kt 8.

As an end game, this is rather a sad exhibition for two masters. The redeeming feature of the game is Rubinstein's fine combination in the middle game, beginning with 14 B × Kt.

GAME 3. IRREGULAR DEFENCE

(Havana, 1913)

White: D. Janowski.          Black: J. R. Capablanca.

At last Black is on his way to obtain full development. The idea of this irregular opening is mainly to throw White on his own resources. At the time the game was played, the system of defence was not as well known as the regular forms of the Queen's Pawn openings. Whether it is sound or not remains yet to be proved. Its good features are that it keeps the centre intact without creating any particular weakness, and that it gives plenty of opportunity for deep and concealed manœuvring. The drawback is the long time it takes Black to develop his game. It is natural to suppose that White will employ that time to prepare a well-conceived attack, or that he will use the advantage of his development actually to prevent Black's complete development, or failing that, to obtain some definite material advantage.

A very serious mistake. I considered castling, which was the right move, but desisted because I was afraid that by playing 13 B × Kt, P × B; 14 Kt - Kt 3, B - Kt 3; 15 Kt - B 5, White would obtain a winning position for the end game. Whether right or wrong this shows how closely related are all parts of the game, and consequently how one will influence the other.

Not good. The natural and proper move would have been Kt - K 3, in order to bring all the Black pieces into play. B × Kt at once was also good, as it would have relieved the pressure against Black's King's Pawn, and at the same time have simplified the game.

Here it is seen how failure to comply with the elementary logical reasons, that govern any given position, often brings the player into trouble. I was no doubt influenced in my choice of moves by the fear of B - B 5, which was a very threatening move.

P - K R 4, to be followed by P - Kt 4, might have been a more vigorous way to carry on the attack. Black's weak point is unquestionably the Pawn at K 4, which he is compelled to defend with the King. The text move aims at doubling the Rooks, with the ultimate object of placing one of them at Q 6, supported by a Pawn at Q B 5, Black could only stop this by playing P - B 4 which would create a "hole" at Q 5; or by playing P - Kt 3, which would tie the Black Queen to the defence of the Q B P as well as the K P, which she already defends. Black, however, can meet all this by offering the exchange of Rooks, which destroys White's plans. For this reason P - K R 4 appears the proper way to carry on the attack.

This move is preparatory to P - K Kt 3, which would make Black's position secure. Unfortunately for Black, he did not carry out his original plan.

P - K Kt 3 would have left Black with a perfectly safe game.

A very serious mistake, which loses a Pawn. P - K Kt 3 was the right move, and would have left Black with a very good game. In fact, if it should come to a simple ending, the position of the Black King would be an advantage.

This wins a Pawn, as will soon be seen. Black cannot reply 24...Kt - K 3; because 25 P × P ch, Kt × P; 26 Q - R 4 wins the Knight.

The game went on for a few more moves, and, there being no way to counteract the advance of White's two passed Pawns, Black resigned.

GAME 4. FRENCH DEFENCE

(St. Petersburg, 1913)

White: J. R. Capablanca.          Black: E. A. Snosko-Borovski.

This constitutes the McCutcheon Variation. It aims at taking the initiative away from White. Instead of defending, Black makes a counter demonstration on the Queen's side. It leads to highly interesting games.

At the time this game was played the variation 5 P - K 5 was in vogue, but I considered then, as I do now, the text move to be the stronger.

This is considered superior to P × P. It has for its object, as I said before, to take the initiative away from White by disrupting White's Queen's side. White, however, has more than ample compensation through his breaking up Black's King's side. It might be laid down as a principle of the opening that the breaking up of the King's side is of more importance than a similar occurrence on the Queen's side.

The plan of Black in this variation is to post his Bishop on the long diagonal so as to be able later on, in conjunction with the action of his Rooks along the open K Kt's file, to make a violent attack against White's King. It is, of course, expected that White will Castle on the King's side because of the broken-up condition of his Queen's side Pawns.

An original idea, I believe, played for the first time in a similar position in a game against Mr. Walter Penn Shipley, of Philadelphia. My idea is that as there is no Black Bishop and because Black's pieces have been developed with a view to an attack on the King's side, it will be impossible for Black to take advantage of the apparently unprotected position of White's King. Two possibilities must be considered. Firstly: If Black Castles on the Queen's side, as in this game, it is evident that there is no danger of an attack. Secondly: If Black Castles on the King's side, White begins the attack first, taking advantage of the awkward position of Black's Queen. In addition to the attacking probabilities of the text move, White in one move brings his King into safety and brings one of his Rooks into play. Thus he gains several moves, "tempi" as they are called, which will serve him to develop whatever plan he may wish to evolve.

Unquestionably a mistake, overlooking White's fine reply, but a careful examination will show that White already has the better position.

This move has been criticised because it puts the Knight out of the way for a few moves. But by forcing Q - K Kt 4; White gains a very important move with P - B 4, which not only consolidates his position, but also drives the Queen away, putting it out of the game for the moment. Certainly the Queen is far more valuable than the Knight, to say nothing of the time gained and the freedom of action obtained thereby for White's more important pieces.

In such positions it is generally very advantageous to get rid of the Black Bishop controlling his Q R 3 and Q B 3, which form "holes" for White's pieces. The Bishop in such positions is of very great defensive value, hence the advantage of getting rid of it.

White threatened P - B 6 ch.

To prevent the Knight from moving to Q 6 via Q 2 and K 4 or Q B 4. It is self-evident that White has a great advantage of position.

I had considered R - Kt 3, which was the right move, but gave it up because it seemed too slow, and that in such a position there had to be some quicker way of winning.

Kt - K 4 or Kt - Kt 3 would have brought about an ending advantageous to White.

If R - Kt 1; 28 Kt × P, R (Kt 1) - B 1; 29 Kt × P would win.

Kt - B 4 was the right move. I was, however, still looking for the "grand combination," and thought that the Pawn I would later on have at Q 6 would win the game. Black deserves great credit for the way in which he conducted this exceedingly difficult defence. He could easily have gone wrong any number of times, but from move 22 onwards he always played the best move.

The position is most interesting. I believe I lost here my last chance to win the game, and if that is true it would vindicate my judgment when, on move 28, I played P - B 4. The student can find out what would happen if White plays Q - Q 4 ! at once. I have gone over the following variations: 35 Q - Q 4, R × K R P (of course if R × B P, P - Q 8 wins); 36 Q × Q P ! R - Q 1; 37 Q - R 6, K - Kt 1 best (if Q - Q 5 ch; K - R 1, K - Kt 1; R - Q Kt 1 wins); 38 Q × B P and White will at least have a draw.

There is nothing to be done against this simple move, since White cannot play Kt - Q 4, because Q - R 8 mates.

A very interesting battle.

GAME 5. RUY LOPEZ

(St. Petersburg, 1914)

White: Dr. E. Lasker.          Black: J. R. Capablanca.

The object of this move is to bring about speedily a middle-game without Queens, in which White has four Pawns to three on the King's side, while Black's superiority of Pawns on the other side is somewhat balanced by the fact that one of Black's Pawns is doubled. On the other hand, Black has the advantage of remaining with two Bishops while White has only one.

Black's idea is to Castle on the King's side. His reason is that the King ought to remain on the weaker side to oppose later the advance of White's Pawns. Theoretically there is very much to be said in favour of this reasoning, but whether in practice that would be the best system would be rather difficult to prove. The student should notice that if now all the pieces were exchanged White would practically be a Pawn ahead, and would therefore have a won ending.

A perfectly sound form of development. In any other form adopted the Black Kt could not be developed either as quickly or as well. K 2 is the natural position for the Black Kt in this variation, in order not to obstruct Black's Pawns, and also, in some eventualities, in order to go to K Kt 3. There is also the possibility of its going to Q 5 via Q B 3 after P - Q B 4.

This move I considered weak at the time, and I do still. It leaves the K P weak, unless it advances to K 5, and it also makes it possible for Black to pin the Kt by B - Q B 4.

Best. It threatens B - B 4; B - K 3, Kt - Q 4. It also prevents B - K 3 because of Kt - Q 4 or B 4.

Preparatory to P - Q Kt 3, followed by P - Q B 4 and B - Kt 2 in conjunction with Kt - Kt 3, which would put White in great difficulties to meet the combined attack against the two centre Pawns.

It has been wrongly claimed that this wins the game, but I would like nothing better than to have such a position again. It required several mistakes on my part finally to obtain a lost position.

Played against my better judgment. The right move of course was B × B. Dr. Lasker gives the following variation: 13...B × B; 14 R × B, P - B 4; 15 Q R - Q 1, B - Kt 2; 16 R - B 2, Q R - Q 1; 17 R × R, R × R; 18 R - Q 2, R × R; 19 Kt × R, and he claims that White has the best of it. But, as Niemzovitch pointed out immediately after the game, 16...Q R - Q 1 given in Dr. Lasker's variation, is not the best. If 16...Q R - B 1 ! then White will have great difficulty in drawing the game, since there is no good way to stop Black from playing Kt - B 3, followed by Kt - K 4, threatening Kt - B 5. And should White attempt to meet this manœuvre by withdrawing the Kt at Kt 3; then the Black Knight can go to Q 5, and the White Pawn at K 4 will be the object of the attack. Taking Dr. Lasker's variation, however, whatever advantage there might be disappears at once if Black plays 19...Kt - B 3, threatening Kt - Kt 5 and also Kt - Q 5, neither of which can be stopped. If White answers 20 Kt - Q 5, Kt - Q 5 for Black will at least draw. In fact, after 19...Kt - B 3 Black threatens so many things that it is difficult to see how White can prevent the loss of one or more Pawns.

It is a curious but true fact that I did not see this move when I played 13...B - Kt 2, otherwise I would have played the right move 13...B × B.

The game is yet far from lost, as against the entry of the Knight, Black can later on play P - B 4, followed by P - Q 4.

I now was on the point of playing P - B 4, to be followed by P - Q 4, which I thought would give me a draw, but suddenly I became ambitious and thought that I could play the text move, 17...Kt - B 1, and later on sacrifice the exchange for the Knight at K 6, winning a Pawn for it, and leaving White's K P still weaker. I intended to carry this plan either before or after playing P - K Kt 4 as the circumstances demanded. Now let us analyse: 17...P - B 4. If 18 Kt - Q 5, B × Kt; 19 P × B, P - Q Kt 4; and a careful analysis will show that Black has nothing to fear. Black's plan in this case would be to work his Kt around to K 4, via Q B 1, Q Kt 3, and Q B 5 or Q 2. Again, 17...P - B 4; 18 R - B 2, P - Q 4; 19 P × P, B × P; 20 Kt × B (best, since if R (B 2) - Q 2, B × Kt give Black the advantage), R × Kt; 21 R × R, Kt × R; and there is no good reason why Black should lose.

Once more changing my plan and this time without any good reason. Had I now played R × Kt; P × R ch, R × P; as I intended to do when I went back with the Knight to B 1, I doubt very much if White would have been able to win the game. At least it would have been extremely difficult.

This, of course, has no object now. Black, with a bad game, flounders around for a move. It would have been better to play R - R 6 to keep the open file, and at the same time to threaten to come out with the Knight at Kt 3 and B 5.

Again bad. White's last two moves were weak, since the White King does nothing here. He should have played his Rook to Kt 3 on the 27th move. Black now should have played P - Kt 4 ch. After missing this chance White has it all his own way, and finishes the game most accurately, and Black becomes more helpless with each move. The game needs no further comment, excepting that my play throughout was of an altogether irresolute character. When a plan is made, it must be carried out if at all possible. Regarding the play of White, I consider his 10th and 12th moves were very weak; he played well after that up to the 27th move, which was bad, as well as his 28th move. The rest of his play was good, probably perfect.

GAME 6. FRENCH DEFENCE

(Rice Memorial Tournament, 1916)

White: O. Chajes.          Black: J. R. Capablanca.

Of all the variations of the French Defence I like this best, because it gives Black more chances to obtain the initiative.

Though I consider P × P the best move, there is much to be said in favour of this move, but not of the variation as a whole, which White adopted in this game.

The alternative, P - K Kt 3; leaves Black's King's side very weak. White by playing P - K R 4 would force Black to play P - K R 4; and later, on White's Bishop by going to Q 3, would threaten the weakened K Kt P. By the text move Black gives up Castling, but gains time for an attack against White's centre and Queen's side.

Threatening Q - R 4 and stopping thereby White's threat of B - R 3. It demonstrates that White's last move was a complete loss of time and merely weakened his position.