• Author / Uploaded
• Sergio Cruz Fernández

Citation preview

Kings, Commoners and Knaves Further Chess Explorations Edward Winter

Foreword by Yasser Seirawan

Russell Enterprises, Inc. Milford, CT USA

Kings, Commoners and Knaves Further Chess Explorations

© Copyright 1999 Edward Winter All Rights Reserved ISBN: 1-888690-04-6 Published by: Russell Enterprises, Inc. PO Box 30 Milford, CT 06460 USA http://www.chesscafe.com [email protected] Library of Congress Catalog Card Number: 98-87479 Printed in the United States of America

Except as noted below, all photographs used in this book are courtesy Edward Winter. Signature facsimiles and the photograph of Olga Capablanca Clark courtesy the Russell Collection.

Table of Contents

Foreword

ii

Preface

iii

References

iv

1

I

Positions

II

Games

III

Openings

141

IV

Miscellaneous

169

v

Gaffes

262

VI

M ysteries

303

VII

Reviews

332

VIII

Quotes

375

43

Book List

399

Games and Positions Chronologically

412

Index of Games and Positions by Player

419

Index of Openings

424

General Index

426

Foreword

Like so many others in the chess world, I first became familiar with Edward Winter' s writings through his publication Chess Notes, which was a blend of scholarly chess journalism, historical research, topical material and accurate good riddance for the lazy and sloppy authors of the chess world. The charm of Chess Notes lay in its good humour and down-to-earth approach, bringing chess scholarship down from the ivory tower and making it relevant to virtually all lovers of the Royal Game. His highly successful book Chess Explorations was a compendium from the journal ' s eight-year run, and now, in this same vein, we have Kings, Commoners and Knaves. In this Information Age, when most of the yearly deluge of chess books falls into two or three drearily predictable categories, it is a delight to open a book that contains material of the sort not to be found elsewhere. Winter is also a dedicated foe of the hackneyed untruths that populate so many chess books, and I promise you that you will never look at chess history in the same way after you read this book. That is not to say that history is all that you will find here. Kings, Commoners and Knaves also contains topical commentary and hundreds of games and positions from all levels of play. Above all, this is a chess book to enjoy and to cherish. Combining scholarly and lively writing, it is uncommonly well­ researched and well-written, and has that delightful undercurrent of humour and wit that has always graced Winter' s writings. Kings, Commoners and Knaves is a rare chess book, to be treasured for years to come. Yasser Seirawan Three-time US Chess Champion Seattle, September 1 998

ii

Preface

In this collection of further chess delvings the twofold objective remains to offer fresh, reliable material and to rectify various misconceptions, and worse, which continue to permeate the chess world. It is a world of kings, commoners and knaves, all forms of life from the sublime to the repellent. The kings, of course, are the game ' s foremost luminaries, whose glorious and tragic careers are the very cornerstone of chess history. The commoners include a large cast of unsung heroes dedicated to the game and meriting our recognition and respect. Last, and certainly least, come the knaves, relatively small in number, often incompetent, but withal pernicious. Heartfelt thanks are extended to readers for their continuing support of our sleuthing and, in particular, to the correspondents from some 20 countries whose contributions are included in the present book. Thanks are also due to the various magazines which have given permission for material to be reproduced here. Technical advice and proofreading assistance have been provided, far beyond the call of duty, by Mr Richard Forster (Winterthur, Switzerland) and Mr Jonathan Manley (liford, England). Finally, special thanks must be offered to the publisher, Mr Hanon Russell, for his exceptional enthusiasm and expertise throughout the production process. Readers' comments, however critical, may be communicated to the publisher and will be warmly welcomed. Edward Winter Geneva, December 1 998

iii

References

The C.N. number is specified in the case of items which have appeared in the Chess Notes series, the publication schedule of which has been: C.N.s 1-291: C.N.s 292-592: C.N.s 593-871: C.N.s 872-1071: C.N.s 1072-1299: C.N.s 1300-1521: C.N.s 1522-1788: C.N.s 1789-1933:

1982 1983 1984 1985 1986 1987 1988 1989

C.N.s 1934-2020:

C.N.s 2117-2145:

1993 1994 1995 1996

C.N.s 2146-2181:

1997

C.N.s 2182-2187:

1998

C.N.s 2188-

1998-

C.N.s 2021-2075: C.N.s 2076-2116:

C.N.s 1934-2187 appeared in a syndicated column run by the author. Thereafter C.N. became a regular column exclusively in New in Chess. Other previously published writings are identified by the year of publication, preceded by one of the following abbreviations: A

American Chess Journal

B

British Chess Magazine (BCM)

C

CHESS

CC

The Chess Cafe (Internet)

1

Inside Chess

K

Kingpin

N

New in Chess

S

Schacknytt.

iv

I Positions

Find the win

A superb example of how every detail of a position needs to be examined. It would seem immediately obvious that White mates quickly by 1 �xh7+ 'it'xh7 2 l:! h3+ 'it'g6 3 li:Je7 mate. Or l li:Je7+ 'it'h8 2 �xh7+ 'it'xh7 3 l:! h3 mate. Or l li:Jf6+ gxf6 2 gxf6+ 'it>h8 3 �xh7+ xh7 4 l:! h3 mate. Standard mating combinations ­ except that there is no mate. All the above variations are impossible owing to the pin on the a8-h 1 diagonal. The winning line is simply l li:Je7+ and 2 �xb7. Sourc e : Wiener Schachzeitung, M arch 1 9 2 3 , page 1 4 . The position arose in a 1 907 game; Krej c ik was White, his grandmaster opponent being unnamed. (741 & 796) Unusual finishes

Three unusual finishes: (See diagram, top of next page.) White, to move, found the quickest conclusion: 26 'it>d2 , with three ways of mating on the following move. Source: tournament book, pages 42-43 .

H. von Gottschall-C. Walbrodt, Kiel, 1 September 1 893 .

J.R. Capablanca-J. Grommer, New York, 2 July 1 9 1 3 . White exploited his opponent' s back-rank weakness b y 3 9 fl e8 fl f4 40 �b8 �g8 41 'lillb 3+ �h8 42 fl xf8+ fl xf8 43 'lillf7 'lilieS 44 'lill x f8+ Resigns. Source: American Chess Bulletin, August 1 9 1 3 , page 1 7 1 .

A . Brinckmann-R. Keller, B ad Oeynhausen, 1 8 July 1 939. White administered the coup de gra ce as follows: 27 fl d8 fl xd8 28 fl h7+ Resigns. It is mate in two. A rare double rook sacrifice. Source: Lachaga tournament book, pages 38-39. 2

(C 1985)

Instant mates

Mate in one move.

Solution: There needs to be a white square in the bottom right-hand comer. Turning the board 90 degrees in one direction allows 1 �d4 mate and, in the other, 1 fxg8(.£l) mate. This problem by Alapin was published on page 88 of the March-April 1 9 1 6 Wiener Schachzeitung, together with two easier mate-in-one positions (based respectively on the familiar tricks of the en passant rule and nine black pawns).* (1 158 & 1 1 95) The 'Christmas Puzzle' hereunder, by J.C.J. Wainwright, appeared on page 262 of the December 1 9 1 7 American Chess Bulletin:

'White mates in one move. '

The solution was given on page 50 of the February 1 9 1 8 issue. It must be Black' s move; whatever he plays, White then mates in one move. A position to ponder

Three posers (See diagram, top of next page.): a) White to move and win. b) Where has this position been widely seen? c) Where (1497) has it been printed before? a) The winning combination is easy enough: 1 �xb4 axb4 2 § xa8+ Ae8 3 1txd5 . *A problem by W. Langstaff in which White has nine pawns and mates in one after the removal of any one of them was given on page 554 of The Social Chess Quarterly, July 1 935.

3

b) The position was played out in an episode of the NBC television series Columbo in which the world chess champion murdered his closest rival. c) It is given, for instance, on pages 1 68- 1 69 of Winning Chess by Irving Chemev and Fred Reinfeld. (1557) Congratulations to Jack O' Keefe, who has discovered that the position did come from a real game: W.J. Wolthuis-C.H.O'D. Alexander, Maastricht, 1 946. Nimzo-l ndian Defence. 1 d4 fl :B.xf2 + 1 2 'it>e 1 � fl + 1 3 'it>xfl �h4 1 4 �b4+ d6 1 5 �xd6+ cxd6 1 6 cxd4 (2166) Ah3+ 1 7 'it>g1 �e 1 mate. From Richard Forster: 'The announcement is correct, but 13... '!fh4 ? allows 14 cxd4 and White escapes immediate mate, though not defeat. The correct line is 13... '!fg5!, i. e. 14 '!fb4+c5 15 '!fxc5+.ti.xc5 16 d4 '!fg2+ 17 r!!le1 '!fe2 mate. ' A Reshevsky problem

The following problem, composed by Reshevsky before he was nine years old, was one of two published on page 1 82 of the November 1 920 American Chess Bulletin :

Mate in three moves. 1 'it>f4 f2 ( l . . .h2 2 'it>e3 f2 3 .ilxf2 mate) 2 Ah2+ 'it>xh2 3 .£\f3 mate.

10

(K 1992)

Pomar problems

Compositions by notable players are always of interest. The following mate-in­ three problems were composed by the Spanish prodigy Arturo Pomar (born 1 93 1 ) when he was about ten years old, according to page 23 1 of La vida de Arturito Pomar by Juan M. Fuentes and Julio Ganzo.

1 .§ a4 �f5 2 E! g8, or l . . .�d5 2 E! c8, etc.

1 &L:!a4 bxa4 2 Aa3, etc.

(1945)

Double queen sacrifice

Few games have two consecutive queen sacrifices, but one example, involving double promotion, is K. Gilg v K. Lamprecht, played at Karbitz on 1 8 August 1 924 :

11

Black played 68 . . .fl (�) 69 �xfl h 1 (�) 70 �xh1 Stalemate. Source: Osterreichische Schachrundschau, September 1 924, page 270. (1948) Capablanca fragments

Compilers of 'complete games' volumes are often undecided about whether to include game fragments. Here is one that is seldom seen anywhere:

J.R.Capablanca (simultaneous)-N.N., Moscow, 1 935. With Black to move, play went 1 7 . . . �h5 1 8 Ae2 �h4 1 9 Axg4 Axg4 20 .§ d4 .§ xe3+ 2 1 'it>fl .§ e 2 22 f8 2 �b4+ Ae7 3 �xe7+ xe7 4 f8(�)+ .§ xf8 5 Ag5+

.§f6 6 exf6+ gxf6 7 .ll xf6+ 'it>f7 8 .llh 4+ 'it>g6 9 .§ g4+ 'it>h5 10 .§ l f4 e5 1 1 .§ g5+ 'it>h6 1 2 .§ f6+ �xf6 1 3 .§h5+ 'it>xh5 1 4 Axf6 e4 and the game was agreed drawn

a few moves later. Source: 64, 31 1 935, inside front cover. 12

(K 1993)

Critical moments

Pages 234-237 of Chess with the Masters by Martin Beheim give the game R. Fine v P. Keres, A VRO, 1 93 8 . After 39 moves this position arose:

The American' s move 40 'ifilxel is criticised as inferior to 40 §.xel, which, it is claimed, would have drawn. Then Beheim comments: 'The consequences of a tiny slip like this may have affected the whole of modern chess history . For Fine tied with Keres in thi s great A VRO tournament, but the application of the tie-breaking system gave Keres the official right to challenge for the world championship. Had Fine been the winner, it is quite possible that his American compatriots would have found the backing for a title match with Alekhine - and Fine would have had excellent chances against an Alekhine who was, in 1 938, well past his peak. It is scarcely exaggerating to suggest that Reuben Fine came within one move of becoming the first world champion from the United States. What actually happened was that Fine, disappointed by not winning the AVRO tournament and his inability to dislodge Reshevsky from the United States championship, gradually withdrew from international chess.' We have lost count of the fallacies in this fanciful speculation about the consequences of a game played as early as round seven out of 1 4. Can readers quote better examples of decisive positions in chess history? (1958) Tarrasch studies

Siegbert Tarrasch is seldom thought of as a study composer, but two of his compositions were published on page 309 of the 1 92 1 Kagans Neueste Sc hac hnac hric hten:

13

White to move and win.

1 'it>f6 'it>h5 2 g8(�) �xg8 3 'it>g7 'it>g5 4 h3 'it>h5 5 h4 and wins. Or l . . .'it>h4 2 g8(�) �xg8 3 f8 3 �d6+ 'it>e8 4 �b5+ 'it>f7 5 �c4+ 'it>g6 6 �d3+ and to avoid mate or loss of the queen Black must go back to f7, allowing perpetual check. If 6 . . . �h6 7 .IUB+ �h5 8 g4+ �h4 9 ..lle7+ �h3 10 ..llf1 mate. Tarrasch regarded 'study' as an unsuitable term; he preferred 'winning (or drawing) problems ' , as opposed to 'mating problems ' ( The Game of Chess, page 77). (K 1993) Alekhine queen sacrifices

From Alekhine in Europe and Asia by John Donaldson, Nikolay Minev and Yasser Seirawan here are four little-known queen sacrifices from simultaneous games: (See diagram, top of next page.) 32 E! xb5 �a6 (if 32 ... cxb5 then 33 c6 mate) 33 ..llc3 �xa3 34 E! b7+ �dB 35 �xc6 mate.

14

A. Alekhine-L. Litov, Serpukhov, November 1 9 1 5 .

A. Alekhine-F. Segovia, Buenos Aires, 1 9 August 1 926. 26 tfxe6 �c6 (or 26 . . . dxe6 27 Ae5+) 27 Ae5 + �xe5 28 tfxe5+ Resigns.

A. Alekhine-Maurer, Vienna, 4 October 1 930. 25 4Jxe8 .§. xc2 26 ff6+ 'JJg7 27 .§. xd7 .§. c l + 28 'JJh 2 tfb6 29 4Jg5 'JJh6 30 4Jxf7 + 'JJg7 3 1 4Jd6+ 'JJh6 32 .§. xh7+ 'JJg 5 33 4Jfe4+ 'JJf4 34 .§. h4 mate.*

*The book gave '33 d7 2 2 �g5 .§ df8 23 .§a2 'it>e7 24 .§ af2 f6 25 �f3 g5 26 �e1 .§ t7 27 b5 .§ a8 28 �c2 .§ ff8 29 �b4 g4 30 h4 �g8 31 �d5+ �d7 32 �xf6+ � xf6 33 .§ xf6 .§ xf6 34 .§ xf6 .§ a3 35 .§f5 .§ xb3 36 .§ xh5 .§ b2 + 37 'it>fl .§b1 + 38 �e2 .§ b2 + 39 �d1 .§ b 1 + 40 �c2 .§g1 4 1 .§ h7+ �d8 42 h5 .§ xg3 43 h6 .§h3 44 �b3 g3 45 'it>c4 g2 46 .§ g7 .§ xh6 47 .§ xg2 .§ h3 48 .§ g8+ �d7 49 .§ g7+ �d8

(See diagram, top of next page.) 50 �d5 .§ xe3 51 �c6 .§ xd3 52 .§ d7+ 'it>e8 53 .§ xc7 d5 54 exd5 e4 55 �d6 e3 56 c4 'it>f8 57 .§ e7 .§ c3 58 .§ e4 e2 59 .§ xe2 .§ xc4 60 �e6 'it>e8 61 d6 .§ d4 62

59

� a2 � e4+ 63 'it'd5 � e 1 64 'it'c6 � c l + 65 'it'xb6 'it'd7 66 � d2 �b1 67 'it'a6 � a 1 + 68 'it'b7 �b1 69 b6 �b3 70 � d5 �b2 71 'it'a7 � a2+ 72 'it'b8 � b2 73 b7 �b3 74 'it'a7 'it'c6 75 b8(4J)+ 'it'xd5 76 d7 Resigns.

Source: tournament book, pages 25 1 -252. Readers sometimes pass over lengthy games, but nobody should miss the above finale. Underpromotion in a rook ending is a very rare occurrence. (N 1 988) An unpublished Kasparov game

H.E. Ballo offers this game: G. Kasparov (simultaneous)-H.E. Ballo, Cologne, 29 October 1 988. Queen 's Indian Defence. 1 d4 4Jf6 2 c4 e6 3 g3 c5 4 4Jf3 b6 5 Ag2 Ab7 6 d5 exd5 7 4Jh4 g6 8 4Jc3 .llg7 0-0 0-0 1 0 .llg 5 h6

1 1 Axf6 Axf6 1 2 4:lxg6 Axc3 1 3 bxc3 � e8 14 4:lf4 4Jc6 1 5 Axd5 �g5 1 6 e4 � ad8 1 7 4Jd3 4Ja5 18 f4 �g7 19 e5 Aa6 20 �a4 d6 21 � ae 1 dxe5 22 4:\xe5 � e7 23 4:\ xf7 � xf7 24 � e8+ � xe8 25 �xe8+ �f8 26 Axf7+ 'it'g7 27 �xf8+ 'it'xf8 28 Ad5 4:lxc4 29 �e1 4Ja3 30 � e6 .lld3 3 1 � xh6 4Jb5 32 c4 4Jc3 33 'it'f2 'it'g7 34 � c6 4:lxa2 35 � c7+ 'it'f6 36 g4 'it'g6 37 h4 4Jb4 38 h5+ 'it'h6 39 'it'g3 4:lxd5 40 cxd5 Resigns.

The position in the diagram arose in Capablanca v Marshall, Carlsbad, 1 929, 60

which continued 1 1 4Jxd5 . The tournament book points out that List-Siimisch, Berlin, 1 927, had proceeded 1 1 ..ll xf6 ..llxf6 1 2 4Jxg6. Analysis of this line also appeared on pages 40-45 of Tartakower' s Das neuromantische Schach as well as on page 54 of Alekhine' s book on the 1 927 New York tournament. It would be interesting to learn how much of all this was known to Kasparov when he (K 1 990) played 1 2 4Jxg6. Ten seconds per move

David Hooper and Dale Brandreth' s The Unknown Capablanca published two of the Cuban ' s rapid transit games (against Meyer and Rosenthal), 'all that could be found although Capablanca was the fastest player in the world for more than twenty-five years' (page 1 00). Jack O' Keefe has now discovered a third specimen, played in the finals of a ten­ seconds-per-move tourney held j ust before the New York, 1 924 international tournament. Since Capablanca finished first with 8-2, a point ahead of Schapiro and Tenner, the game had a decisive influence on the final standings. M.A. Schapiro-J.R. Capablanca, New York, 1 1 March 1 924. Queen 's Gambit Declined. 1 d4 d5 2 4Jf3 4Jf6 3 c4 e6 4 4Jc3 c6 5 e3 4Jbd7 6 Ad3 b6 7 0-0 ..llb7 8 e4 dxe4 9 4Jxe4 c5 10 'i!¥e2 cxd4 1 1 4Jxd4 Ac5 1 2 4Jb3 0-0 13 ..llg5 Ae7 1 4 .§ ad 1 4Jxe4 1 5 ..ll x e7 'i!¥xe7 16 ..ll x e4 ..ll x e4 17 'i!¥xe4 4Jf6 18 'i!¥h4 .§ ac8 19 .§ d4 aS 20 .§ fd1 a4 2 1 4Jd2 e5 22 .§ d3 'i!¥b4 23 b3 axb3 24 axb3 .§ fd8 25 4Je4 .§ xd3 26 4Jxf6+ gxf6 27 Resigns.

Source: Christian Science Monitor, 1 5 June 1 924. Jack O' Keefe comments : 'If the score is correct, Schapiro resigned when he had a simple perpetual check by 27 'lii'g4+, and Capablanca could have won by 26.. : itJ/8 (27 fJxh7+ �e8). Not easy to see at ten seconds per move. ' A crosstable of the event was published on page 52 of the March 1 924 American Chess Bulletin. The field included Mar6czy and Tartakower. (K 1 990) Castling trick

A further contribution from Jack O' Keefe: 'After the Manhattan Chess Club International of 1 948-49, Euwe visited Detroit for a few days. He stayed at the home of a friend who one evening inv ited some local players over to meet the former world champion. Euwe

61

graciously agreed to play some casual games with us, one of which was a charming miniature. ' M. Palmer-M. Euwe, Detroit, 1 949. Sicilian Defence. 1 e4 cS 2 ff3 a6 3 d4 cxd4 4 4)xd4 4)f6 S Ad3 eS 6 4)b3 4)c6 7 0-0 dS 8 exdS 4)xdS 9 Ae4 Ae6 10 �f3 �d7 1 1 t'!d1 Ag4 12 t'! xdS Axf3 13 t'! xd7 Axe4 1 4 t'! xb7

14 . . . 0-0-0 1 S Resigns.

A novel example of the familiar trick of queen' s-side castling to win a stray rook at Kt2. Other cases can be found on pages 1 1 - 1 4 of Tim Krabbe' s book (K 1990) Chess Curiosities. Robert John McCrary sends the following game featuring the same manreuvre: R..J. McCrary-S. Wassner, Correspondence game. Albin Counter-Gambit. 1 d4 dS 2 c4 eS 3 dxeS d4 4 4)f3 4)c6 S 4)bd2 Ag4 6 h3 Ae6 7 a3 aS 8 �b3 t'! a6 9 e3 AcS 10 exd4 t'! b6 1 1 �d3 4)xd4 1 2 4)xd4 .ilxd4 1 3 4)f3 .ilxb2 14 �xd8+ �xd8 1 S Axb2 t'! xb2 16 0-0-0+ Resigns.

Proposals for a name for this castling trick would be welcome - something graphic (K 1993) rather than jokey. We have now found a much earlier specimen: G.H. Thornton-Boultbee, 1 884 (?). King 's Gambit Declined. 1 e4 eS 2 f4 AcS 3 4)f3 d6 4 d4 exd4 S 4)xd4 4)f6 6 4)c3 0-0 7 .ile3 �e7 8 .ild3 t'! e8 9 a3 4)g4 1 0 Ag1 fS 1 1 Ae2 fxe4 1 2 4)dS �f7 1 3 Ac4 Ae6 14 4)xe6 t'! xe6 1 S �xg4 c6 1 6 �xe6 �xe6 1 7 4)e7+ 'M8 1 8 Axe6 �xe7 1 9 Ac8 4)d7 20 Axb7 § b8 21 .ilxc6 § xb2

(See diagram, top of next page.) 22 Axd7 �xd7 23 AxeS dxcS 24 0-0-0+ Resigns.

62

Source: Brooklyn Chess Chronicle, November 1 884, page 3 1 . Jeremy Gaige' s Chess Personalia lists George Howard Thornton (born in Watertown, NY on 28 April l 85 1 , died in Buffalo, NY on 30 January 1 920). Unless an earlier game can be found, 'Thornton castling trap' might be an appropriate (K 1994) term. Below is another game position which confutes Tim Krabbe' s belief (see Chess Curiosities, page 1 1 ) that this attractive castling combination was unknown until Selesniev' s 1 92 1 study.

P. Krii ger-Stein, Hamburg, May 1 908. Play went: l . . . § xb2 2 dxe5 dxe5 3 �a8+ �dB 4 �xd8+ �xd8 5 0-Q-0+ and wins. Source: Deutsches Wochenschach, 2 May 1 909, page 1 5 3 . *

(K 1 996)

King hunt W . C. Spencer-N.N., Fort Snelling (USA), 1 884 (?). Four Knights ' Game. 1 e4 e5 2 �f3 �f6 3 �c3 �c6 4 Ac4 �xe4 5 Axf7+ �xf7 6 �xe4 d5 7 �fg5+ �g6 8 �f3 dxe4 9 �f7+ 'it>xg5 1 0 d3+ 'it>h4 1 1 g3+ 'it>h3 12 �h5+ �g2 13 Ag5 �e8

(See di agram, top of next page.) *This queen ' s-side castling ruse was subsequently discussed by Robert Timmer in Chapter 3 o f his book Startling Castling!

63

Source: Brooklyn Chess Chronicle, November 1 884, page 3 1 .

(K 1994)

Capablanca's games

At the Central Cafe in Vienna on 2 December 1 9 14, Richard Reti won a miniature against Dunkelblum ( 'of Cracow' ) : R . Reti-Dunkelblum, Vienna, 2 December 1 9 14. Three Knights ' Game. 1 e4 e5 2 g8 1 9 exf6 Resigns. (A computer check shows that after 1 9 . . . .£)f5 White would have given mate in, at most, eight moves.)

S ource: The Spirit of the Times, 1 4 July 1 860. J.A. Leonard-L. Mark, New York. (Date?) Scotch Game. 1 e4 e5 2 .£)f3 .£)c6 3 Ac4 h6 4 d4 exd4 5 0-0 d6 6 .£)xd4 .£)e5 7 Ab3 c5 8 f4 .£)c6

9 '?!!h 5 g6 1 0 '?!!d 5 '?!!c7 1 1 .£)b5 '?!!d7 1 2 e5 .£)b4 13 .£)xd6+ Axd6 14 '?!! x d6 '?!! x d6

1 37

1 5 exd6 .ile6 1 6 a3 .ilxb3 1 7 axb4 Axc2 1 8 bxc5 f5 1 9 b4 .:£lf6 20 .:£lc3 a6 2 1 E! e 1 + Ae4 2 2 Ab2 E! f8 2 3 b 5 �d7 2 4 .:£la4 .:£ld5

25 c6+ bxc6 26 .:£lc5 + �xd6 27 .ila3 E! fe8 28 .:flxe4+ �d7 29 bxc6+ �xc6 30 E! ac l + �b6 31 Ac5+ �c7 32 Aa7+ �b7 33 .:£ld6+ �xa7 34 .:flxe8 .:flxf4. It is mate in five (35 E! c7+ �b6 36 E!b1 + �a5 37 E! c2, etc.).

Source: The Dial, 4 October 1 86 1 . Moreover, thanks to Mr Jack O' Keefe, via Mr Pope, the following four games may be added. They appeared in Hazeltine' s 1 866 book Brevity and Brilliancy in Chess (on pages 3, 45 , 82 and 94-95 respectively) . [J.?] Thompson-J.A. Leonard, Occasion? King 's Gambit Accepted. 1 e4 e5 2 f4 exf4 3 .:£lf3 g5 4 .ilc4 .ilg7 5 d4 d6 6 c3 c6 7 ii1b3 ii1e7 8 .:flxg5 i?1xg5 9 .ilxf7+ �f8 10 .ilxg8 E! xg8 1 1 0-0

Black mated in two. J.A. Leonard-Jos. Leonard, Occasion? (Remove White ' s rooks at a 1 and h l .) 1 d4 d6 2 c4 e5 3 e3 c5 4 d5 .:£lf6 5 .:£lc3 .ilg4 6 f3 .ilh5 7 .:£lb5 .:£le4 8 ii1a4 .:£lf6 .

White mated in two moves. The score was published by Claudius Hiither (Schnell Matt!, page 1 00) and by Kurt Richter (666 Kurzpartien, page 1 0 - he called it 'a text-book example of the power of double check'), without any date, giving 'J.A. Leonhard' [sic] as White and with no identification of B lack, who was James' brother, Joseph. The 138

latter was described by Hazeltine (on page 368 of the December 1 88 1 Brentano 's Chess Monthly) as 'a nice, good-natured, rollicking boy, vehemently devoted to three laudable objects - Muggins, the Morphy Chess Rooms, and his brother' . J.A. Leonard-F. Perrin, Occasion? Petroff Defence. 1 e4 eS 2 �f3 �f6 3 Ac4 �xe4 4 �c3 �f6 S �xeS dS 6 Ab3 Ad6 7 d4 0-0 8 AgS h6 9 Ah4 Ae6 1 0 f4 cS 1 1 "ii1'd2 c4 1 2 Aa4 a6 1 3 0-0-0 bS 1 4 fS AxfS l S .§ dfl Ah7 1 6 .§ xf6 gxf6 1 7 �xdS AxeS 1 8 dxeS �d7 1 9 "ii1' x h6 fxeS 2 0 Axd8 .§ axd8 2 1 .§ fl fS 22 �e7+ �h8 23 �g6+ �g8 24 .§ f3 .§f7 2S t=i g3 f4. White

mates in five. J.A. Leonard-Jackson, Occasion? (Remove White' s queen.) 1 e4 eS 2 �f3 �c6 3 Ac4 AcS 4 c3 �f6 S d3 �g4 6 d4 exd4 7 h3 dxc3 8 hxg4 cxb2 9 Axb2 d6 1 0 �bd2 �xg4 1 1 0-0-0 �d4 1 2 .§ de l �xf3 1 3 gxf3 Ae6 1 4 ..lld3 �xa2 l S eS d S 1 6 e 6 f6 1 7 f4 Ab4 1 8 �f3 Axel 1 9 Ag6+ �e7 2 0 Aa3+ xe6 21 �d4+ �d7 22 MS + �e8 23 .§ xel + �f7 24 Ae6+ �g6 2 S .§gl + �h6.

It is now mate in four (and not five, as stated in Hazeltine' s book). The final two games have been supplied by Mr Jeremy Gaige. Culled from the Macon Telegraph of 1 867, both are from eight-board blindfold exhibitions. J.A. Leonard (blindfold)-E.W. Bryant. New York, 25 October 1 86 1 . Scotch Gambit. 1 e4 eS 2 d4 exd4 3 �f3 AcS 4 Ac4 �c6 S c3 �f6 6 0-0 dS 7 exdS �xdS 8 cxd4 Ab6 9 �gS MS 1 0 �c3 �ce7 1 1 �xdS �xdS 1 2 .§ e l + Ae6 1 3 �xe6 fxe6 1 4 'l!YhS + �f8 1 S .§ xe6 �f6 1 6 "ii1'f3 "ii1' xd4 1 7 Ah6 "ii1' x f2 + 1 8 "ii1'xf2 Axf2 + 1 9 �xf2 �g4+ 20 �g3 �xh6 2 1 .§ fl + �fS+ 22 .§ xfS+ �g8 23 .§ e8 mate. J.A. Leonard (blindfold)-C.A. Gilberg. New York, 1 7 January 1 862. Ruy LOpez. 1 e4 eS 2 �f3 �c6 3 AbS a6 4 Aa4 bS S Ab3 d6 6 0-0 ..lle 7 7 d4 exd4 8 AdS "i!:Yd7 9 �xd4 ..llb7 1 0 �c3 �f6 1 1 �xc6 Axc6 1 2 Axc6 "ii1' x c6 1 3 �dS "ii1'd7 1 4 f4 �xe4 l S .§ e l fS 1 6 �xe7 "ii1' x e7 1 7 "ii1'd S .§ c8 1 8 "ii1' xfS �d8 1 9 .§ xe4 'l!i'f8 20 "i!:Yg4 h6 21 fS dS 22 t=i e6 c6 23 Af4 'lli'f7 24 f6 gxf6 2S .§ d6+ Resigns.

It is recorded that the entire latter display lasted under three hours. The most blindfold games Leonard ever played simultaneously was apparently ten ( +4 -4 =2) in New York on 16 November 1 86 1 . His opponents included Derrickson. Leonard came onto the chess scene just as Morphy was easing himself out of it. On page 262 of his biography of Morphy, David- Lawson stated that in New York in October 1 860 'Morphy met J.A. Leonard and Otto Michaelis, offering both the queen' s rook' , but no further details are recorded. Leonard played most of his chess in New York, although he did visit Philadelphia in 1 86 1 for a match against William Dwight ( 1 83 1 - 1 888), who was later to become a General. With Leo nard leading by +6 -3 =2 and requiring only one more win, the match was 1 39

left unfinished and he returned to New York. Seguin was not alone in ranking Leonard among America' s very best. According to page 8 of Reichhelm' s Chess in Philadelphia (published in 1 898), Leonard 'was, after Morphy, the most promising player America ever produced' . But then came the Civil War, and Leonard was gone, at the age of 20. The circumstances were reported by Hazeltine in the Macon Telegraph of 1 867: 'Moved in an evil hour - by what mocking friend we know not - he enlisted on 1 February 1 862 to Company F., 88 N.Y. Volunteers, an Irish Regiment. He was in the battle of Fair Oaks, and in the seven days' battles, till that of Savage ' s Station where he was captured. Though detained less than three months, so ill would his frame bear the unavoidable hardships that he was attacked with fever and scorbutic dysentery and died at Annapolis. The sad intelligence o f hi s death was conveyed t o the writer by his brother, who with his now doubly bereaved mother.repaired to Annapolis to soothe his last moments. But before reaching him, on 26 September, his spirit departed. He remembered with blessings his friends in his last hours. ' I n Brentano 's Chess Monthly, December 1 88 1 , page 369, Hazeltine recalled that Leonard ' s 'favourite pupil' , the above-mentioned E.W. Bryant, 'never tires of recording Leonard' s games, or of taking lessons from him, or of penning his praises in his letters' and that 'the chess world is indebted to him for the score of many of his young master' s games, which would otherwise have been lost' . It must be hoped that more of Leonard ' s brilliancies will come to light. Incredibly enough, chess players today are most unlikely to find a single game by him in their books or, even, in those million-game databases. That is a gross injustice ( CC 1 998) which has to be rectified. John S. Hilbert draws our attention to this game, played by a seven-year-old against his father: W.D. Brereton-Master Brereton, Pittsburg, 1 898. Steinitz Gambit. 1 e4 e5 2 �c3 �c6 3 f4 exf4 4 d4 �h4+ 5 e2 d6 6 00 ltg4 7 Axf4 0-0-0 8 h3 .ilxf3+ 9 �xf3 'll1tf6 10 �e3 g5 1 1 Ag3 Ah6 12 �d5 g4+ 13 Af4 Axf4+ 14 �xf4 gxh3 15 gxh3 § e8 16 h4 �ge7 17 Ah3+ �b8 18 .ilg2 § hg8 19 �e2 M+ 20 d2 § xg2 21 exf5 �xd4 22 �e1 il1te5 23 Resigns.

Source: Philadelphia Public Ledger, 1 0 February 1 898, which contained Kemeny' s annotations. He wrote that Black was probably 'the youngest chess player living' .

1 40

III Openings

The Griinfeld Defence

'One of the most remarkable Griinfelds ever played ' is how Jack O ' Keefe describes the following game: 1 d4 lLlf6 2 c4 g6 3 lLlc3 d5 4 e3 Ag7 5 lLlf3 0-0 6 cxd5 CL\xd5 7 Ae2 CL\xc3 8 bxc3 c5 9 0-0 cxd4 10 cxd4 CL\c6 1 1 Ab2 ltg4 1 2 § c l § c8 1 3 Aa3 t1a5 14 t1b3 § fe8 1 5 § c5 t1b6 1 6 § b5 t1d8 1 7 CL\g5 Axe2 1 8 CL\xf7 CL\a5 and White mates in three.

Why remarkable? Because the game was played not just before Griinfeld' s career but even before Morphy' s . According to the Chicago Tribune of 1 3 July 1 879, which took the score from the Glasgow Herald, the game was played between Cochrane and Moheschunder in May 1 85 5 . It was also annotated by Steinitz in The Field of 6 June 1 874. (K 1 990) Kasparov, Karpov and the Scotch

We present some background notes on the variation in the Scotch Game played twice by Kasparov against Karpov in the 1 990 world championship match. In particular, it will be shown how, 1 1 0 years ago, the same line was part of a bitter analytical dispute between the world' s leading active players, Steinitz and Zukertort. 1 e4 e5 2 4)f3 4)c6 3 d4 exd4 4 4) xd4 4)f6 5 4) xc6 bxc6 6 e5 tflle7

Tarrasch (in The Game of Chess) approved of 6 . . :rtfe7, but earlier master comment had been negative. In the English-language book of the Hastings, 1 895 tournament, Albin said that it was 'not good', the usual continuation being 6 . . . CL\d5 (a move which Tarrasch considered unsatisfactory) . In The Modem Chess Instructor, published i n 1 889, Steinitz had given 6 . . . t1e7 a question mark, with the following analysis: 7 t1e2 CL\d5 8 c4 Aa6 9 f4 0-0-0 1 0 141

�f2 �b4+ (or 1 0 . . . �b6 1 1 c5 -'1.xf1 1 2 cxb6 -'1.a6 1 3 bxc7, with the better game) 1 1 -'1.d2 �xb2 1 2 cxd5 Axf1 1 3 �xfl �xa1 (or 13 . . 5.!tb7 1 4 -'1.c3 Ab4 1 5 Axb4, with advantage) 14 �a6+ �b8 15 �e2 �b2 16 �c3, with the better game. Another Steinitz line was 9 . . . �b4+ 10 �d1 �b6 1 1 b3 �c8 12 Ad2 �b6 1 3 �c3 Ae7 14 �h5, again with the better game for White.

7 �e2 4)dS

8 c4

Several commentators have called this the 'Mieses Variation' apparently because Jacques Mieses occasionally adopted it in the 1 890s. But, as will be seen below, Blackbume used it twice against Zukertort in match play in 1 8 8 1 , at which time Mieses' chess career had barely begun. Tarrasch (op. cit.) said that 8 c4 was bad because of 8 . . .-'1.a6 9 f4 �b4+ 1 0 �d1 (if 10 �d2 �xf4) 10 . . . Ac5 , after which one possibility, he said, would be 11 a3 �b3+ 1 2 �c2 �e3+ 13 Axe3 �xe3 14 �d2 d6 1 5 exd6 0-0-0, ' and Black has the far superior game' . Instead of 8 c4, Tarrasch recommended 8 �d2 Ab7 9 �b3 0-0-0 1 0 c4 �b6 1 1 Ad2 .§. e8 1 2 f4 f6, 'and Black has the superiority in the centre' . An attack by 1 3 a4 would be beaten back by 1 3 . . . Aa6 1 4 �e4 �f7 1 5 c 5 -'1.xf1 1 6 cxb6 axb6 1 7 a 5 b5 . 'All these variations are exceptionally trenchant and lively. ' Up to 16 cxb6 this was the tenth match game Mieses v Tarrasch, Berlin, 1 9 1 6. Tarrasch played the inferior 16 . . . �xb3. 8 �d2 was the move chosen by Blackbume in a blindfold simultaneous game

against Piper in London in October 1 88 1 ( The Chess Monthly, June 1 882, pages 306-307). Play continued: 8 . . . g6 (as good as 8 . . . a5, said the magazine) 9 �f3 Ag7 1 0 a3 ( 'White dare not play 1 0 Ag5 at once, for Black would win with 10 . . . �b4+ at least a pawn ' ) 10 . . . a5 1 1 Ag5 �e6 1 2 c4 �b6 1 3 M4 Aa6, and Black eventually won (at move 40) . Another continuation is 8 b3 a5 9 Ab2 a4 1 0 �d2 axb3 1 1 axb3 .§. xa 1 + 1 2 Axa1 �a3 1 3 �d1 Ab4 (Mieses v Marco, Hastings, 1 895, a game won by Black in 54 moves). Albin (in the English-language tournament book) said: 'White should play 8 c4 (if 8 . . . �b4+ 9 �d2 �f4 1 0 �xb4 Axb4+ 11 Ad2 1 42

.Q.xd2+ 1 2 4Jxd2 4Jg6 1 3 4Jf3), with the better game; and if instead of check with the queen, 8 . . . 4Jb6, then White would get a better development of his game.'

In the same tournament book, Schiffers called 8 f4 'good continuation here' . Mieses v Forgacs, Ostend, 1 907 went 8 f4 f6 9 c4 Aa6 10 b3 fxe5 1 1 fxe5 0-0-0 1 2 Ab2 'l!i'g5 1 3 4Jd2 .1lb4 1 4 Ad4 .§ de8 1 5 a 3 l=! xe5 1 6 .1lxe5 .1lxd2+ 1 7 �d1 Ac3 1 8 .§ c l .ll x e5 19 'l!i'f2 4Je3+ 20 �e2 Ad4 21 'l!i'f3 Ab7 22 'l!i'g3 'l!i'e7 23 �d3 c5 24 �d2 4Jf5 25 'l!i'h3 Ae3+ 26 Resigns. s

..

. Jta6

Karpov' s choice in the 1 4th game, 1 990. The match book by Kasparov, Geller, Lein and Chepizhny says (page 1 00) that Karpov' s choice in the 1 6th game, 8 . . . 4Jb6, 'is not new, and had already occurred in tournament play ' . That is quite an understatement, given that 8 . . . 4Jb6 even appeared in Bilguer 's Handbuch in the nineteenth century. lt offered the variation 8 . . . 4Jb6 9 M4 Aa6 1 0 4Jd2 'l!i'b4 1 1 0-0-0 .§ b8 1 2 a3 'l!i'a4, which prompted The Chess Monthly to observe that 'some of the moves for White show more valour than discretion' . Another line: 8 . . .-l!i'b4+ 9 4Jd2 4Jf4 1 0 'l!i'e4 4Je6 1 1 f4 Ab7 1 2 a3 'l!i'b6 1 3 4Jf3 c5 (Walbrodt v Marco, Hastings, 1 895, a 58-move draw). In the English-language tournament book, Schiffers said that 9 �d1 was preferable. A modem instance of Marco ' s line of development (except that Marco castled on the king ' s side) was praised a s ' quite a neat concept' b y Miles o n page 5 o f Inside Chess, 29 December 1 990. 9 b3

For 9 f4, see Schiffers' notes to the correspondence game Chardin v Schiffers, given on page 252 of the English edition of the Hastings, 1 895 tournament book. Also Deutsches Wochenschach, 1 895, page 8 3 . Various spellings of Chardin ' s name are to b e found. 9 ...0-0-0

In his 1 4th match game against Karpov at Lyon in 1 990, Kasparov played 10 g3 now, and this caused players with the black pieces to examine other choices at move nine (9 . . :�·h4, 9 . . . g6 and 9 . . . g5). Here we continue with the old line. 10 .Q.b2

One of the most often quoted games is Mieses v Teichmann, Hastings, 1 895, 1 43

which continued 10 Ab2 4Jb6 1 1 g3 t! e8 1 2 Ah3 f6 1 3 0-0 fxe5 and was won by Black at move 53.* 1 0 Ab2 was also played in the 1 4th game of a match in London between Blackbume and Zukertort in 1 88 1 , a match which Zukertort won with a score of +7 -2 =5 . Steinitz ' s annotations in The Field were strongly criticised by his enemies Zukertort and Hoffer, who were the joint editors of The Chess Monthly. An acrimonious analytical controversy on the complete set of 1 4 match games took up countless pages of these and other journals.

10... �g5

This was Zukertort' s reply to B lackbume' s 10 Ab2, and, in a rare instance of harmony, both Steinitz and Zukertort/Hoffer believed it to be B lack' s best move . * * This 1 4th match game continued (after 1 0 Ab2 �g5) as follows: l l �e4

Steinitz wrote that 1 1 h4 would be better because if 1 1 . . .Ab4+ then 1 2 'itld1 �g6 13 �c2 with advantage. The Chess Monthly gave 1 1 h4 �g6! * * * l l . .Q.b4+ 12 �d1 4)e7 13 h4 ..

The Chess Monthly calls this the only move, parrying the threatened 1 3 . . . d5 . It regards 13 f4 as far inferior because of 1 3 . . . d5 14 cxd5 �g4+ 1 5 �f3 �xf3+ (and not 15 ... t! xd5 + 1 6 'itlc2, winning a piece) 1 6 gxf3 Axf1 1 7 § xfl 4Jxd5 . 13 ... �g6

Steinitz says that this is best since the queen would have stood badly at h6, allowing White to gain time for development with 14 'itlc2, which prevents 14 . . . d5 because of 15 exd6 § xd6 1 6 c5 . The Chess Monthly notes the variation 13 . . . d5 14 cxd5 �h5 + 1 5 g4 § xd5 + 16 'itlc2 and wins. 14 �xg6 hxg6 15 �c2

Steinitz describes this as unnecessary and prefers 15 Ad3 , guarding against the knight' s entry at f5 . If then 1 5 . . . d5, White can play 16 exd6, followed by 1 7 'itlc2 and 1 8 t!dl . The Chess Monthly says that the reply t o 1 5 Ad3 would have been 1 5 . . . g5 . * Gewinnen mit Schottisch by Gutman gives 13 e6! dxe6 14 h 1 or 1 8 t'l f2, then 1 8 . . . t'l xe2 1 9 �xe2 �xc3) 1 8 . . . xe2 b7 1 8 t'lg1 �h4+, and Black will be a pawn ahead . * * 1 4 Ad2 d4 15 {)a4 {)d5

According to The Chess Monthly, 15 . . . c5, followed by 16 . . .-'tb7 would be much better. 16 �f3 {)b4 The Chess Monthly wrote: ' 1 6 . . . h2 4Jd2 3 1 §gl e2 32 '1Wxd8+ 'l!txd8 33 § xd2 'l!txd2 34 4Jxd2 §xg2+ 35 § xg2 el('IW) 36 §e2 'l!thl + 37 'it>g3 'l!tgl + 38 'it>h4 g5+ 39 c,t?h5 gxf4 40 §e5 and Black mated in three moves. Source: La Strategie, 15 November 1898, page 344. Typically complex Janowsky play, although various short-cuts were possible (3 l . . . § xg2+, 33 . . . -l!th4 and

39 . . . '1Wg3).

Janowsky was White in this position from a game at the Manhattan Chess Club, New York. He played 1 4Ja6+ c,t?a8 2 4Jxc7+ 'it>b8 3 4Ja6+ 'it>a8 4 §b7 e2 5 §b8+ § xb8 6 4Jc7 mate. Source: American Chess Bulletin, March 1918, page 69. A specimen of his play in a consultation game :

D. Janowsky-Dobell, Mackeson and Watt, Hastings, 1 September 1897. Ruy LOpez.

1 e4 e5 2 4Jf3 4Jc6 3 -'tb5 4Jf6 4 o-o 4Jxe4 5 §el 4Jd6 6 4Jxe5 -'te7 7 Ad3 4Jxe5 8 § xe5 0-0 9 4Jc3 M6 1 0 §e3 g6 1 1 b3 4Je8 1 2 .ila3 d6 1 3 4Jd5 Ag5 1 4 f4 .ilh6 1 5 Ab2 Ae6 16 'l!tf3 c6 1 7 4Jc3 d5

18 f5 d4 19 fxe6 fxe6 20 'l!th3 Axe3+ 21 dxe3 dxc3 22 �xe6+ 'h8 23 Axc3+ 4Jg7 24 §dl �e8 25 �h3 t=!d8 26�h6 l:'=!d7 27 h3 b5 28 l:'=!d2 'it>g8 29 e4 'l!te7 30 c,t?hl 'l!tc5 31 Ab2 4Jh5 32 'h2 t=!df7 33 Ae2 4Jf4 34 Ag4 '1Wb4 35 t=!d4 '�!tel 3 6 l:'=!d7 t=!xd7 37 Axd7 t=!f7 38 Ae8 '1Wxe4 39 .ilxf7+ 'xt7 40 'l!txh7+ 'e8 4 1 'l!th8+ 'd7 4 2�d4+ �xd4 4 3 .ilxd4 a 6 4 4 g4 'e6 4 5 h4 and White won. 259

Source: BCM, October 1 897, pages 395-396. Janowsky played four games simultaneously against teams of three players. Now a sample of Janowsky' s annotations to one of his games:

G. Marco-D. Janowsky Sixth match game, 13 June 1 904. King 's Gambit Declined.* 1 e4 e5 2 f4 Ac5 3 'iif"f3 ( ' An innovation meriting consideration. ' * * ) 3 . . . 4Jc6 4 c3 d6 5 Ac4 4Jf6 ('5 . . . Ae6 was stronger. ' ) 6 d3 0-0 ( 'Premature, in view of the ease with which White sets up a king' s-side attack. Here too 6 . . . Ae6 would be much better. ' ) 7 f5 ( 'Hemming in the enemy queen' s bishop and threatening g4, followed by h4. ' ) 7 . . . d5 8 �b3 ('8 �xd5 4Jxd5 9 exd5 4Je7 1 0 g4 'iit" x d5, etc. and 8 exd5 e4 9 dxe4 4Je5 10 'iif"e 2 4Jxc4 1 1 'iit" x c4 4Jxe4, etc. would be to White ' s disadvantage. ' ) 8 . . . dxe4 9 dxe4 'iit" d6 10 Ag5 ( 'With 10 g4 he could have put his opponent in a very tricky position. ' ) 10 . . . h6 1 1 �xf6 ( 'The reply to 1 1 �h4 would have been 1 l . . . .§d8.') 1 l . . . 'iit" xf6 1 2 4Jd2 .§d8 ( 'Taking possession of the open file and having in mind the possibility of being able, if necessary later on, to escape with the king via f8 and e7. ' ) 1 3 0-0-0 a5 ( ' To set up a counter-attack and chase the bishop away from the very awkward diagonal which it occupies. ' ) 14 h4 ( ' In positions of this kind there is rarely time to defend, and victory goes to the player who can attack the fastest. ' ) 14 . . . a4 15 �c2 b5 16 4Je2 b4 17 g4 bxc3 ( ' Too soon. 17 . . . �a6, threatening 1 8 . . . Axe2, followed by 19 . . . bxc3, would have been much stronger. ' ) 18 4Jxc3 ( 'Forced, for if 18 bxc3 then 18 . . . �a3+, etc . ' ) 18 . . . 4Jd4 1 9 'iit"g 2 a3 20 g5 ('20 b3 would be more solid. ' ) 2 0 . . . axb2+ 2 1 �b1 'iit"c6 22 gxh6 ( 'He has n o time to play 22 f6 because of 22 ... Ab4 . ' ) 22 . . . 'iit" xh6 23 4Jb3 �3 24 4Jb5 ( ' A tempting move, the consequences of which White did not examine sufficiently closely . ' )

2 4 . . . 4Jxb5 ( 'The attack h e obtains provides ample compensation for the sacrifice of the exchange . ' ) 25 .§ xd8+ �h7 26 .§ d3 ( 'If 26 'iif"g3 then 26 . . . .§ b8, threatening mate in a few moves by 27 . . . 'iit"c 1 +. After 27 .§dd1 , Black wins with 27 . . . 'iit" c 6. ' ) *This was the final game of a match (won by Janowsky witb a score of +4 -2 =0) which was played on board S.S. Pretoria as Marco and Janowsky returned to Europe after their success at the Cambridge Springs tournament. During their journey out to America, on the same ship, eight masters (including Marco and Janowsky) had played a consultation game, tbe score of which is given on pages 1 1 1 - 1 1 2 above. **In fact, 3 'iol'f3 was played twice by Charousek in 1 896, against Showalter and Mar6czy.

260

26 . . . �b8 27 �g5 ( 'He was threatened with 27 . . :�cl + 28 � xc1 bxcl(�)+ 29 �xcl