Late Knight No. 26
More About Potter
William N. Potter (1840-1895) was introduced in my May column. This month I will concentrate on his role in a famous controversy involving the first two World Champions, and some examples of his play. To start with, another miniature win by Potter against one of the strongest players of his day:
Zukertort - Potter
10th Match Game, London, 1875
1.e4 e6 2.d4 d5 3.exd5 exd5 Curiously, the Exchange Variation was White’s main weapon against the French Defence for several decades… See also the January column for Exchange variation motifs with gxf3/gxf6.
4.Nf3 Nf6 5.Bd3 Bd6 6.0–0 0–0 7.Nc3 Nc6
Nowadays a more flexible setup with knights on d2 and d7 is more often played.
Steinitz, in his notes for The Field, preferred the development of the bishops to e3 and e6 respectively. Several games of Blackburne’s are relevant to this variation. For example, against Schwarz at Vienna 1873, he tried 8...Ne7?!, but after 9.Bxf6 gxf6 10.Nh4! Ng6 11.Qh5 c6 12.Ne2 White obtained a large plus, and eventually won in 32 moves. (See Diagram)
Greedily going after the d-pawn, but in return Black obtains too many attacking possibilities. In later games, White relied mostly on 9.Kh1, though usually without gaining any opening advantage. One example is Berger – Blackburne, from the 1881 Congress of the German Chess Federation in Berlin: 9.Kh1 Be7 10.Be3 Qd7 11.Qd2 (allowing the typical doubling of the f-pawns) Bxf3 12.gxf3 Qh3 13.Be2 Bd6 14.f4 Ne7 15.Rg1 Ng6 16.Rg5 Kh8 17.Rag1 Rae8 18.R1g2 Rg8 19.Bd3 c6 20.Ne2 Ne4 21.Bxe4 dxe4 22.R5g4 Qh5 23.Rg5 Qh3 24.R5g4 Qh5 25.Rg5 Qh3 26.R5g4 Qh5 27.Rg5 Qh3, and a draw was agreed.
In his notes Steinitz recommended 9.Nxd5 Bxh2+ 10.Kxh2 Qxd5 11.Bxf6 gxf6 12.Be2 Bxf3 13.Bxf3 Qxd4 14.Bxc6! with a promising ending, despite the pawn deficit; but later praxis revealed that Black is not forced to enter this line. Instead, 12...Rfe8 gives him a decent position. Two examples went 13.c3 Qh5+ (13...Re4 also seems good) 14.Kg1, and now:
A) 14...f5 15.Qd2 Re6 16.Bd1 Rg6 17.Qf4 Re8 18.Qh2! (Simple, but efficient. White liquidates into a favourable endgame.) Qxh2+ 19.Kxh2 Nd8 20.Ne5 Rh6+ 21.Kg1 Bxd1 22.Rfxd1 Ne6 23.Re1 Ng5 24.Nd3 Re4 25.f3 Reh4 26.Re8+ Kg7 27.Rae1 Rh1+ 28.Kf2 Rxe1 29.Nxe1 Rd6 30.Re5 Kf6 31.Nd3 Ne6 32.f4 b6 33.b4 c6 34.a4 Nc7 35.Ke3 Nd5+ 36.Kd2 Rd8 37.c4?? (Not the sort of tactical mistake that would go unnoticed by Blackburne.) Nxf4! 38.Nxf4 Rxd4+ 39.Ke3 Kxe5 40.Nd3+ Rxd3+ 0:1 (Sellman-Blackburne, London 1883).
B) 14...Ne7 15.Re1 Ng6 16.Qd2 Rxe2! (A very promising exchange sacrifice.) 17.Rxe2 Bxf3 18.gxf3 Nh4? (Tempting, but after the preparatory 18...Kh8! White would have been in real trouble.) 19.Qf4! Nxf3+ 20.Kg2 Nh4+ 21.Kf1 Nf3 22.Re3! Qh1+ 23.Ke2 Qxa1 24.Kxf3, and White had an advantage, which he eventually converted to a win in Chigorin-Berger, Berlin 1881.
Finally, note that if 9.h3, Black can play 9...Bxf3 10.Qxf3 Nxd4 with a clear conscience because unlike in the game (with reversed colours) the square h3 is not available to White’s queen. Rosenthal-Blackburne, Paris 1878, continued 11.Qd1 Ne6 12.Bxf6 Qxf6 13.Qh5 g6 14.Qxd5 with equality.
9...Qxf6 10.Nxd5 Qh6 11.h3? (See Diagram)
Another, already decisive mistake. With 11.Qc1 Qxc1 (11...Qh5 12.Nf4!) 12.Raxc1 Bxf3 13.gxf3 Nxd4 White could keep his disadvantage within bearable proportions. The text-move meets with an instant mating attack.
A desperate attempt to divert the bishop (12.hxg4 Nxf3+ and 13...Qh2 mate), but it is already too late. 12.Be2 (or 12.Be4) would not have saved White either, for Black wins with 12...Nxf3+ 13.Bxf3 Bxh3! 14.gxh3 Qxh3 15.Re1 Bh2+ 16.Kh1 Bg3+, and mate in two moves.
Incidentally, eight years after the present game this line was still unfamiliar to a leading master. In Englisch-Blackburne, London Tournament 1883, White fell into the same trap as Zukertort in this game. He deviated from the above line with 14.Re1, but being a pawn down in a miserable position, he only lasted five moves longer: 14...Be6 15.g3 Rad8 16.Qe2 Bxd5 17.Bxd5 Bxg3 18.Be4 Rd2 19.Qxd2 Bh2+ 0:1.
Proper opening preparation has always come in handy, even 117 years ago …
A last attempt to stir up some confusion was 13.Nf5!?, although Black keeps a clear extra piece after 13...Bxf3.
13...Bxd1 14.Nef5 Qf6 15.Raxd1 Be5 16.c3 Rad8 17.f4 Bxd4+ 18.Nxd4 c5 19.Nb3 Rxd3! 0:1.
A fitting final stroke.
The Sixth Reason
After last month’s five reasons for remembering Potter, here is another context in which some readers may have come across the name of this little-known and modest player and journalist. In his famous Manual, the Lasker devotes many pages to explaining the essence and development of Steinitz’ playing style, which so revolutionised the theory and perception of chess at the time.
Along with many other people over the last few decades, I sometimes wondered what exactly Steinitz’ oft-quoted theory actually consisted of. In vain I checked the text-books and biographies – even in the International Chess Magazine, which Steinitz edited from 1885 until 1891, I did not find the theory. Reading Lasker’s Manual did not really help either, but left me more confused than ever. In the meantime I have had the opportunity to study a few relevant sources.
Unfortunately, the question of what Steinitz’ theory consisted of, and of whether it should not rather be called Lasker’s theory instead, because it was he who formulated the ideas, is outside the scope of this article. However, as it has been addressed in print before, I refer the reader to these publications:
- Purdy’s ‘The Great Steinitz Hoax’ in the Australian Chess Player’s Quarterly 1978, vol. 2 no. 3. Reactions appeared in 1980 and 1981, nos. 1 and 2 of vol. 4 and no. 1 of vol. 5.
- Hooper’s ‘The Theory of Steinitz’ in The British Chess Magazine September 1984.
- Ree’s ‘Accepted Myths’ in New in Chess, 7/1990, with readers’ reactions in issues 1/1991 and 2/1991.
To summarise the above sources, my personal conclusion is that much confusion arose because of different interpretations of what a theory is. I believe that we should think of Steinitz, Paulsen and perhaps some others as the main inventors and discoverers of many important new principles of positional play which distinguish the modern school from its predecessors. Unlike his fellow ‘modernists’ Steinitz, however, also had the gift of being able to formulate those principles and make them known to a wider public through his voluminous and erudite writings. In my opinion, speaking of ‘Steinitz’ Theory’ is justified insofar as we talk of a (coherent) set of concrete chess principles. Lasker, on the other hand, despite being one of the most practical players of all time, had a different outlook on chess and life. He was not satisfied with being an excellent chess player, but sought understanding far beyond the chequered board. His philosophical works are obvious proof of that. In pursuit of his objective of linking chess and life, he created his own much more abstract theory of fight and balance, which is documented in his Manual and other books. For one reason or another he assumed, or at least argued, that these considerations originated with Steinitz, but there is no evidence that the latter actually thought along these lines. Thus, although there is such a thing as Steinitz’ theory, Lasker’s interpretation thereof is something different and should rather be named ‘Lasker’s Theory’.
Perhaps I have managed to explain something or perhaps you are even more confused now. In any case, where does Potter come in? The answer is to be found on page 200 of Lasker’s Manual of Chess (1932):
‘I heard in London, that a London master, Mr Potter, who loved unusual and strange moves, had influenced Steinitz greatly. The two were friends, and Steinitz somehow began to copy Potter’s style. However that may have been, I can well believe that a strange style would rise, almost of necessity, at a time so romantic, so superstitious as that time was. Potter probably saw through the emptiness and the presumption of the style then dominating and with his style of play he seemed to call out to his contemporaries: ”You want to beat me right from the start by force of your greater genius? Look! I make ridiculous moves, and yet you cannot beat me. Become, I pray you, more modest and more reasonable.” ’
It has been pointed out several times since that when Potter and Steinitz first met in London around 1872-3, the foundations of Steinitz’ new style had long since been laid. The first to make this observation was probably B. Goulding Brown in a book review in the British Chess Magazine, February 1933, page 69. After quoting Lasker’s first two sentences regarding Potter he remarked:
‘It would be interesting if Dr. Lasker’s informant, or informants, would tell us more about this. I have always understood that Potter was Steinitz’s pupil, at any rate in the sense that he of all the amateurs in London entered most fully into Steinitz’s views, and was most affected by them. I have certainly read that he alone of the playing committee could understand Steinitz’s plan of campaign in the great correspondence match London v. Vienna, 1872-4, and that the other members gradually ceased to attend, leaving Steinitz and Potter to conduct London to victory. There may, however, be something of legend in this. Not many men could find the time for analysis of two such games, one (the Scotch, defended by London with 4...Qh4) of the most laborious kind.’
In the November 1933 issue we find a reply by Lasker to Goulding Brown and other correspondents, but unfortunately, the Potter matter only being a side issue, he did not explain his reasons for giving the little-known player such high credit.
Without doubt, Potter was on better terms with Steinitz than were most other players of that era, and he also had quite an independent chess view. For a game more typical of his style than the one given at the beginning of this column, see Tim Harding’s Kibitzer column, where another win of Potter’s from that match is analysed.
Commenting on the match against Zukertort, the Westminster Papers wrote that Potter played overmuch with an eye on the endgame, and ‘absorbed in his search for things remote, sometimes overlooked the blessings which lay under his nose.’ And Steinitz, one of Potter’s main contributors to The City of London Chess Magazine, once described Potter’s way of playing as follows: ‘Put all the pieces into a hat and shake them out on to the board, and you have Potter’s style exactly.’
As shown elsewhere, Potter and his magazine were closely connected to the City of London Chess Club. From G.A. MacDonnell’s Chess Life-Pictures (1883) we learn that
‘indisputably, to Mr. W. Norwood Potter belongs the honour of having raised the City Club from comparative obscurity to world-wide eminence. His talents as a player and the amiability of his disposition, together with his practical ability as an administrator, attracted to the club about ten years ago not merely a large body of recruits but well-nigh all the most distinguished players in the metropolis. A chess club that desires to achieve a real and permanent success must, like the City, first and above all things cultivate a spirit of corporate fraternity.’
The big quarrels in the club, which had as a consequence the retirement of almost all the leading players from the club in 1875, must have been a tragedy for Potter. Not only did he stop his magazine thereafter but he must also have suffered through his inability to bring the power struggle to a peaceful end. Regarding this and other issues, Steinitz, in the March 1885 International Chess Magazine (page 83), wrote:
‘Some time back a gentleman asked me what I thought of Potter’s analysis. Potterio is not a personage whom I greatly admire. In London chess politics he tried the experiment of bowing all around in the midst of a fighting crowd, and he looked quite astonished when he found himself alternately kicked in the rear by different parties. However, I knew that Potter was otherwise the most conscientious and painstaking chess writer in the British Metropolis, though this would not say much in itself considering into what hands the chess press in London has fallen [those of Leopold Hoffer – Steinitz’ lifelong literary enemy]. But as he is also a very fine player of the modern school, as well as unquestionably the ablest analyst in England next to Zukertort (...), I answered my interrogator that Potter’s analysis was generally reliable and correct, or words to that effect.’
And later on the same page:
‘Some years ago I told Mr. Potter to his face, not just in reference to analysis in general, but respecting chess political affairs in special, that he could see remarkably clear just beyond his own nose. In fairness I am bound to state that he answered promptly: - ”That is more than most other people can do.” However, on investigating his countermining attack to my favorite defense in the Ponziani I found that Potter had seen remarkably clear just beyond Zukertort’s – sixth move in that opening in the following variation: 1.e4 e5 2.Nf3 Nc6 3.c3 d5 4.Qa4 f6 (this is my defense) 5. Bb5 Nge7 6.d3 Bd7. This last is not any move at all. It was played by Zukertort against Rosenthal in their first match game. I condemned it at once in my notes at the time, and recommended 6...Be6. Try again now Potterio.’
If you think that Steinitz was being quite rude about Potter, you should read what he had to say about some other people…
To claim that Potter had a great influence on Steinitz seems grossly exaggerated, but they certainly knew each other well and had similar ideas in many respects.
‘An Hour With W. N. Potter’
was the title of an article by E. Freeborough in the British Chess Magazine, December 1885. With several examples and quotes the author tries to explain Potter’s chess style. While I am not sure whether he really succeeds, let us just follow him on his interesting tour, which seems to be based mainly on analysis published in the Westminster Papers (which, unfortunately, I don’t have to hand).
‘In Chess, says Mr. Potter, ”you are to take nothing for granted. So far from receiving appearances with open arms you are to view them with suspicion, and test their claims thoroughly.” ’
Freeborough then gives a game Potter-Minchin, London 1877, in which the following position occurred (See Diagram):
Now one would expect 18.h4 or perhaps 18.Bc4, but Potter performs a surprising knight manoeuvre first, judging that his kingside needs further support:
18.Nf4!? Qe7 19.Ng2!
Taking care of the temporary hole on h4.
19...Ne6 20.f4 Bd4
More tenacious was 20...Rae8, as from now on White’s attack seems irresistible.
21.Bxd4 Nxd4 22.Rde1 f6 23.c3 Ne6 24.h4 Kh8 (See Diagram)
25.f5 Ng7 26.Nf4
Also very strong was 26.fxg6, because 26...hxg6 27.Nf4 Kh7 is met by 28.e5!
26...gxf5 27.exf5 Qd7 28.h5!
Sealing Black’s fate as 28...h6 fails to a knight check on g6.
‘A move like this shows that Mr. Potter’s careful style is not grounded on timidity. ”Fitz James’ sword is sword and shield.” ’
The move defends h1 and threatens to win at once with 30.Ng6+!
29...Rf7 30.h6 Rg8
A final desperate attempt to create counter-play on the g-file. After 30...Ne8 31.Bc4 Rf8 32.Re7 Rd8 33.Qh5 Black would be disposed of quickly.
Sure of his victory Potter does not want to take the slightest risk. True, 31.Bc4? Nxf5! would have been highly unclear, but either 31.Be4 or 31.hxg7+ Rfxg7 32.Kh2 won material. The textmove also leads to a quick end, however.
31...Ne8 32.Bc4 Rff8 33.Bxg8 Kxg8 34.c4 Qd7 35.Rfe1 Bc6 36.Re7 Qd8 37.g5 (See Diagram)
Later Potter found that 37.Qh5 with the threat of 38.Qg6+! led to a quick mate.
37...Rf7 38.Rxf7 Kxf7 39.g6+ Kg8 40.gxh7+ Kxh7 41.Qg4 1:0.
The next game shows us Potter as an unconventional defender in a highly uncomfortable position (See Diagram):
Stevens - Potter
Despite an extra exchange, Black’s position does not look enviable. Winning a pawn with 28...Bc3 29.Qc2 Bxb4 is much too dangerous because after 30.f6! White develops a strong attack. Potter therefore starts a sensible regrouping of his pieces, without fearing to put his queen on an awkward post.
28...Nf7 29.Nh5 Qh8!
‘The position of my Queen seemed to amuse Mr. Blackburne, and other lookers on, but I liked her better here than at f8. I don’t say I was right.’ (Potter)
30.a3 Re7! 31.h4 Rbe8 32.Rf3 Bc3 33.Qc2 Bf6
Suddenly afraid of 34.f6!?, Potter refrains from the planned 33...Bd4.
34.Rh3 Kf8 35.hxg5 Bxg5 36.Nhf4 (See Diagram)
This looks very threatening since White will be able to put a knight either on e6 or g6. With a carefully calculated combination Potter, however, cuts the Gordian knot:
36...Bxf4! 37.Nxf4 Rxe3!
37...Qa1+ is useless because of 38.Bg1.
38.Ng6+ Kg8 39.Nxh8
Now one intermediate move (39.Rxe3) is met by another (39...Qa1+!).
A strange mistake. After 39...Rxh3+ 40.Kg2 Rhe3 Black will emerge with two rooks and a minor piece for the queen, with an easy win.
Perhaps 40.Qxe2!? Rxe2 41.Nxf7 Kxf7 42.Rxh6 Ne5 43.g5 offered more drawing chances. (See Diagram)
40...Rd2! 41.Rg3 Re1+
Much better than 41...Nfe5 42.Re3 Kxh8 43.Qxd7!
42.Rg1 Ree2 43.Rg3
Going down quickly. Much more stubborn is 43.Rg2! Rxg2 44.Qxd7, although after 44...Rh2+ 45.Kg1 Rdg2+ 46.Kf1 Rf2+ 47.Kg1 Rhg2+ 48.Kh1 Ne5! (not 48...Nxh8 49.f6!) Black should probably be able to escape from White’s checks and achieve a winning position. After the textmove the attack quickly becomes decisive.
44.Qb3 Kxh8 45.g5 Nd3 46.Qxd3
Or 46.Re3!? Rxe3 47.Qc3+ N7e5 48.Qxd2 Rh3+ 49.Kg2 Rh2+! and wins.
Our three final examples, two from the match against Zukertort (+2 –4 =8) and another one against the same opponent from a club tournament, show Potter doing what he did best – defending difficult positions:
Zukertort – Potter
8th Match Game, London 1875 (See Diagram):
Instead of choosing the normal continuation 27...Re5, when Black is certainly not worse, Potter opted for a questionable transformation into an ending with one rook versus two:
27...f4?! 28.Nxd5 Qxd5 29.Qxd5 cxd5 30.Rd2
Of course, such unbalanced positions are very difficult to play, but it seems that 30.Rg4 f3 31.Kf2 would have given White good winning chances.
Stopping 31...f3, which could follow after 31.Rxd5.
Losing an important tempo. After 31...g5 the position should be about equal.
32.Kg2 g5 (See Diagram)
Avoiding the trap 33.Kf3 g4+! and Black is winning!
33...e2 34.Re1 f3+ 35.Kf2 h5 36.Rd5 g4 37.Rg1 Rh7 38.Rh1
As was discovered later, Zukertort could have won here with 38.Rf5!, threatening to take the f-pawn. And if 38...Kh8, simply 39.Rf8+ first.
Securing the draw. Despite having set up a blockade, White cannot win the resulting endgame.
39.Rg5+ Kf7 40.Rgxh5 Rxh5 41.Rxh5 g3+ 42.Ke1 g2 43.Rg5 Kf6 44.Rg8 Ke7 (See Diagram)
White’s pieces are both tied to the three passed pawns, whereas his potential passed pawn on the c-file will be taken care of by the black king. After another 26 moves the game was given up as drawn.
Potter - Zukertort
11th Match Game, London, 1875
1.f4 d5 2.e3 Nf6 3.Nf3 e6 4.Be2 Bd6 5.0–0 0–0 6.d3 b6 7.h3 c5 8.Qe1 Nc6 9.g4 e5 10.Qh4 e4 11.Ne1 Ne7 12.Nc3 exd3 13.cxd3 d4 14.Ne4 dxe3 15.Nxf6+ gxf6 16.Bxe3 Ng6 17.Qg3 f5 18.g5 Re8 19.Nc2 Bb7 20.Rae1 Qc7 21.Bd2 Re7 22.Bd1 Rae8 23.Rxe7 Rxe7 24.h4 c4 25.h5 cxd3 26.hxg6 dxc2 27.gxf7+ Rxf7 28.Bh5 Re7 29.Qb3+ Kf8 30.Qh3 Qc5+ 31.Kh2 Re1!! (See Diagram)
A thunderbolt which might have prompted many a lesser player to resign at once, but Potter finds a way to resist further, and, with a little help from his opponent, even manages to draw miraculously.
Both 32.Rxe1 Qf2+ and 32.Bxe1 c1Q are obviously out of question.
32...Bd5 33.Qd3 c1Q
A simple way to maintain a huge advantage consisted of 33...Rxf1 34.Qxf1 Qd4, whereas the quickest win was 33...Bc4! 34.Qc3 Be5!, when 35.Rxe1 is refuted by 35...Qf2+ 36.Kh3 Bf1+. But this was not easy to calculate under time pressure, and the move chosen seems to win too.
34.Bxc1 Rxc1 35.Qxf5+ Kg8?
But with this move he finally spoils the win. After 35...Ke7 Black is probably still winning.
36.Rxc1 (See Diagram)
This check seems devastating, but upon close inspection it turns out that Black has no path to victory.
37.Kh3 Bg2+ 38.Kg4 Bf3+ 39.Kh3 Qg2+ 40.Kh4 Qf2+ 41.Kh3 Bg2+
These repetitions serve to fulfil the time limit.
42.Kg4 Bf3+ 43.Kh3 Qg2+ 44.Kh4 Qf2+ 45.Kh3 Qg2+ 46.Kh4 Qh2+ 47.Qh3 Qxf4+
‘47...Qxh3 and 48...Bxh5 would but draw.’ (Zukertort).
48.Bg4 Bxg4 49.Qxg4 Draw agreed.
After 49...Qxc1 50.Qe6+ White gives perpetual check. Note that with the king on e7, Black would be winning in the final position.
In our final example, Potter’s ‘piece lumping’ on the back rank is worth comparing to Steinitz’ best attempts in that field. But the real merit of the game lies in the brilliant queen manoeuvre that saves Black at the very last moment:
Zukertort – Potter
Simpson’s Divan Tournament, London 1876
Steinitz did not (yet!) approve of Potter’s strategy in this game when annotating it for his famous column in The Field. We give some of his comments below:
‘We unhesitatingly disapprove of meeting a close opening in this manner. Though Morphy, and also the council of the Viennese in their match against London, have made the attempt, the progress of the examples on record does not justify the imitation.’
‘Already the king’s bishop is shut out from the line where he is usually wanted, and apparently there is no better way of continuing the second player’s game, which is a clear proof that the latter is labouring under greater difficulties of development than he would have done if he had answered ...e6 on the first move.’
3.Nc3 Nf6 4.d4 Be7 5.Nf3 Nc6 6.Be2 0–0 7.0–0 Bd7 8.d5
Doubtful according to Steinitz, who preferred 8.a3.
8...Nb8 9.Nd2 Ne8 10.f4 (See Diagram)
‘By this and the following move the door is opened to the enemy. The pawn at e5 ought to have been well preserved, and Black would have done much better now further to protect it by ...f6. At any rate, the exchange of pawns liberates the square d4 for further occupation by the hostile pieces.’
11.exf4 f5 12.Nf3 g6 13.Bd2 Ng7 14.Qc2 Be8
‘Eccentric style. 14...Na6 presented obviously greater facilities for Black’s development.’
Some fifteen years later this was exactly the sort of comment that annotators would make about Steinitz’ moves!
15.Rae1 Bf7 16.Rf2 Nd7 17.Bd3 Re8 18.Rfe2 Nf8 (See Diagram)
‘Black’s pieces are now packed together in one lump on the king side, while the other wing remains little protected.’
19.Qb3 Rb8 20.Be3 b6 21.Nd4 Qd7 22.Nc6 Ra8
If 22...Rb7, then 23.c5! dxc5 24.Ba6.
Much stronger than 23.Bxb6 according to Steinitz, but the latter move was certainly not so bad either!
23...Bf6 24.Nbxa7 h6 25.Nb5 Kh7 26.Bd4 Rxe2 27.Rxe2 Bxd4+ 28.Nbxd4 Re8 29.Rxe8
Steinitz recommended 29.Qc2, followed by creating a passed pawn on the a-file, as the simplest winning method.
29...Qxe8 30.Kf2 Nh5 31.Nb5 Nd7 32.Qc2 (See Diagram)
Setting the trap 32...Nxf4 33.Bxf5! gxf5 34.Qxf5+ Ng6 35.Nxc7 and wins, but much better was immediate 32.Nxc7 (Steinitz).
‘Another and worse version of the same error. Even now he ought to have captured the c-pawn, for the bishop had always a safe retreat at f1 when twice attacked. By the move in the text an opening is given to the enemy, of which the latter takes clever advantage.’
33...Nxd3+ 34.Qxd3 Nxf4 35.Qf3?
‘Still feeble. By offering the exchange of queens at e3, he could keep victory in his grasp. Supposing 35.Qe3 Qxe3+ 36.Kxe3 Nxg2+ 37.Kf2 Nf4 38.Nxc7, followed by the irresistible march of the a-pawn, while Black’s pieces are easily kept aloof.’
35...g5 36.Nxc7 (See Diagram)
The following manoeuvre, ‘finely conceived’ in Steinitz’ words, is in fact unique in tournament play:
37...Qa1!! 38.Qc2 Qh1!!
A magic triangle. Despite being at the remotest place from her own king, the queen saves Black.
39.Qxf5+ Bg6 40.Qd7+ Kh8 Draw. (See Diagram)
White must give perpetual check. Playing for a win with 41.Qg4? would even lose: 41...Nd3+ 42.Ke3 (or 42.Kg3 Qe1+ and wins the queen) Qg1+ 43.Kd2 Qc1+ 44.Ke2 Qe1+ 45.Kf3 Qf2 mate.
Richard Forster: Late Knight Column, No. 26, ChessCafe.com, June 2000.