I received a couple of Jerome Gambit games from chessfriend Vlasta Fejfar. The first involves a complicated, frustrating, and, ultimately, philosophical defense. The second is almost off-the-road adventuring.
Let's dive into the tough stuff first. After a theoretical opening "discussion" and "scientific" middle game, there follows a textbook attack and a pleasant checkmate.
vlastous - franciscoribeiro
internet, 2017
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
The annoying "annoying defense" (see 1 and 2 for starters). Computers love it. Vlasta has a lot of experience facing it.
Black offers to return one of the two sacrificed pieces. Although his King appears a bit precarious, much of the dynamism in the game is drained off.
White need to go into the line with a plan - and a decision about what kind of outcome he is looking for.
8.fxe5 dxe5 9.Qh3+ Kd6 10.Qd3+ Ke7 11.Qg3
Here the game Fejfar,V - Pressl, corr Czech Republic, 2015 was drawn.
Is a draw acceptable to White, who started the game with a Bashi-Bazouk attack, sacrificing two pieces? Is a draw acceptable to Black, who, only a few moves ago, had an "objectively" won game?
It is a typical Jerome Gambit irony that could be expressed in the idea: Both sides stand better (or worse).
11...Kf8
A little bit better is 11...Kf7, although Vlasta has experience with that line, too: 12.Qxe5 Bd7 (12...Bd4 13.Rf1+ Nf6 14.Qh5+
Kg8 15.Qe2 Be6 16.c3 Be5 17.g3 c5 18.d3 Qd6 19.Bf4 Bg4 20.Qe3 Bh3 21.Rf3
Bg4 22.Rf1 Re8 23.Nd2 Bxf4 24.gxf4 b5 25.e5 Nd5 26.Qg3 Qg6 27.Ne4 c4 28.Kd2 Bf5 29.Nd6 Rd8 30.dxc4 bxc4 31.Rae1 Qxg3 32.hxg3 Ne7 33.Ke3 Bd3 34.Rg1 Nf5+ 35.Nxf5 Bxf5 36.Rd1 Kf7 37.Rd4 h5 38.Rgd1 Rc8 39.Rh1 g6 40.Rh2
Ke7 41.a4 Rc6 42.Rd5 Be6 43.Rb5 Rc7 44.Rd2 h4 45.gxh4 Rxh4 46.Rd4 Rh3+
47.Kf2 Rd3 48.a5 Rxd4 49.cxd4 Bd7 50.Rb8 Ke6 51.Ke3 Kd5 52.a6 c3 53.bxc3
Rxc3+ 54.Kf2 Bc8 55.Ra8 Rc7 56.Ke3 Rc3+ 57.Kf2 Kxd4 58.Rxa7 Ke4 59.Rg7
Kf5 60.a7 Ra3 61.Rf7+ Kg4 62.Rf8 Bb7 63.e6 Rxa7 64.e7 Bc6 65.e8=Q Bxe8
66.Rxe8 Kxf4 67.Rf8+ Kg4 68.Rc8 g5 69.Rc3 Kh4 70.Kg1 Ra2 71.Rb3 g4 draw, Fejfar,V - Goc,P) 13.Qh5+ g6 14.Qxc5 Qh4+ 15.Qf2+ Qxf2+ 16.Kxf2 Nf6
17.d3 Rhf8 18.Nc3 Kg7 19.Ke2 Bg4+ 20.Ke3 Be6 21.h3 Nh5 22.Ne2 Rae8 23.b3
Nf6 24.Ba3 Rf7 25.c4 Rd8 26.Raf1 Rfd7 27.Nf4 Bg8 28.Bb2 Rf8 29.e5 Re7 30.Kd2 Ne8 31.e6+ Black resigned, Fejfar,V - Svoboda, corr Czech Cup, 2016
He has also seen 11...Ke8 12.Nc3 Bd4 13.Rf1 Qd7 14.Nd5 c6 15.Ne3 Nf6 16.d3 Qc7 17.c3 Bxe3 18.Bxe3 Qe7 19.O-O-O
Rf8 20.Rf3 Bd7 21.Rdf1 c5 22.Bxc5 draw, Fejfar - Kyzlink, corr Czech
Republic, 2015.
You may have noticed in some of these games Vlasta was testing the Jerome Gambit in correspondence play - like Alonzo Wheeler Jerome did with his gambit over a century ago. A draw is a reasonable outcome.
12.Qxe5
For comparison, a couple of other games:
12. Rf1+ Nf6 13. Qxe5 Bd6 14. Qg5 Bxh2 15. Nc3 Be6 $2 16. e5 h6 17. Qe3 Qe7
18. exf6 gxf6 19. d3 c5 $6 20. Bd2 Rd8 21. O-O-O b6 $6 22. g3 Kg7 23. Rh1 Rd4
24. Rxh2 h5 25. Re1 Kf7 26. Rhe2 Rd6 27. Ne4 Rc6 28. Qf3 Kg6 29. Ng5 fxg5 30.
Qxc6 Kh7 31. Rxe6 Black resigned, Wall,B - Shah,V, chess-db, 2015; and
12.d3 Nf6 13.Rf1 Qe7 14.Nc3 c6 15.Bg5 Kf7 16.O-O-O Rf8 17.h3 Kg8 18.Rf3 Qe6
19.Rdf1 Be7 20.Kb1 Bd7 21.Nd1 Rae8 22.Qh4 b5 23.Ne3 Qd6 24.Qf2 Be6 25.g4
c5 26.Nf5 Bxf5 27.gxf5 Nh5 28.Bc1 Nf4 29.h4 c4 30.d4 Nh5 31.d5 b4 32.Qe2
c3 33.b3 a5 34.a4 bxa3 35.Rxc3 Black resigned, Wall,B - ubluk, Chess.com, 2012.
12...Qh4+
The alternative, 12...Bd6, was seen in a number of games in the legendary Fisher-Kirshner - KnightStalker match in 1993. When people send me Jerome Gambit games, they usually start with Amateur - Blackburne, London, 1884, and then follow with the Fisher-Kirshner - KnightStalker games.
The text is sharp and relatively unexplored. The only other game with it in The Database is the computer game WB Nimzo 2000b - La Dame Blanche 2.0c, Jerome Gambit thematic tournament, 2009 - which was a 109 move draw!
13.g3 Qe7 14.Qxe7+ Kxe7
The game has left the path of the computers (which contained 14...Nxe7) and has transposed to 3 games played by Philidor1792 in 2012.
15.Nc3
Alternately, d2-d3 was seen in Philidor1792 - NN, 5 0 blitz, 2012 (1-0, 30) and c2-c3 was seen in Philidor1792 - NN, 5 0 blitz, 2012 (1-0, 22) and Philidor1792 - NN, no time control, 2012 (0-1, 27).
15...c6
Played to keep White's Knight off of d5. Probably better was 15...Nf6, but Black seems to have been nervous about a possible Bishop pin at g5 (see move 17).
16.Na4 Bd6 17.d4 h6 18.O-O
Bh3 19.Rf3 Nf6
A puzzling move. Black gives back his extra piece and secures what should be an even position. Stockfish 8, instead, suggests castling-by-hand on the Queenside, 19...Rf8 20.Bf4 Kd8 21.Nc5 Kc8, with advantage.
It must be said that Black is employing the "scientific" idea (as he did on move 7) of accepting the sacrificed material, and then giving it back some time later.
20.e5 Bxe5 21.dxe5 Nd7
Vlasta suggested that 21...Ng4 would have led to an even game.
22.b3
Nxe5
Black is in too much of a hurry to capture the pawn. It will cost him another piece.
23.Re3 Kd6 24.Ba3+ Ke6 25.Rae1 b5 26.Rxe5+ Kf6 27.Bb2 Kg6 28. Nc5 Rad8
White is winning now - he has an advantage in material and a developing attack on the enemy King.
29.Re7 Rhg8 30.Ne6 Bxe6 31.R1xe6+ Kh7 32.
Rxc6 Rd1+
One last Hurrah. White's pressure on g7 is deadly.
33.Kg2 Rb1 34.Bf6 a5 35.Rcc7 Kg6 36.Bxg7 Rd1
37.Rc6+ Kg5 38.Re5+ Kg4 39.h3 checkmate
Chessfriend Vlasta faces the "annoying defense" and shows that while White does not have have a lot to work with, neither does Black. The game is an extended battle, after which the point is split. It is not clear who was satisfied with the result, the defender who nicked his half point against a refuted opening, or the attacker who "survived" playing a refuted opening.
Vlastous - NEWMAN 1982
Internet, 2017
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
So annoying. Black gives back the piece and drains a lot of dynamism out of the position.
8.fxe5 dxe5 9.Qh3+ Ke7 10.Qg3 Ke8
Vlasta has faced 10...Kf7 as well, in Fejfar,V - Goc,P (1/2-1/2, 70); Fejfar,V - Chvojka (0-1, 32); Vlastous 2456 - irinat 2597, ChessManiac.com, 2016 (0-1, 38); and Fejfar,V - Svoboda/corr Czech Cup, 2016 (1-0, 30).
11.Qxe5+
A small improvement over his earlier 11.Nc3 in Fejfar,V - Kyzlink, corr Czech Republic, 2015 (1-0, 20).
11...Qe7 12.Qxe7+ Nxe7
Now we have a battle between the extra two pawns and the extra piece in a Queenless middlegame. White cannot claim an advantage - but how is Black to win?
13.c3 Ng6 14.d4 Be7 15.Be3 Rf8 16.Rf1 Rxf1+ 17.Kxf1 b6 18.g3 Ba6+ 19.Kf2 Bd3 20.Nd2 Kd7
21.Bf4 Rf8 22.Ke3 Ba6 23.Nf3 Bb7 24.h4 h5 25.c4 Rd8 26.Rf1
Re8 27.b3 Bf6
Black's Bishop pair is focused on White's pawn center which has a protected passed pawn.
28.e5 Be7 29.Ng5 Rf8 30.Rf2 Ba3 31.e6+ Kc8 32.Kd3 Nxf4+ 33.gxf4 Kd8 34.f5 Ke7 35.Rf1 Bd6
Black seems unsure what to do. One idea, suggested by Stockfish 8, is to undermine the support of White's advanced passer, while exchanging Rooks, i.e. 35...g6 36.fxg6 Rxf1 37.g7 Rf8 38.gxf8=Q+ Kxf8.
36.Nh7 Rh8 37.Ng5 Rg8 38.Ke3 Bg3 39.d5 Bxh4 40.Kf4 c6
Interestingly enough, this attempt at undermining support is not successful, in that it opens a line for White's Rook. The piece activity allows White to give up a second pawn.
41.d6+ Kxd6 42.Rd1+ Kc5 43.Ne4+ Kb4 44.Rd7 Bc8 45.Rxa7 b5 46.cxb5 cxb5 47.Rc7 Ba6 48.Rf7 Ka5 49.Ke5 Kb6
50.e7 Re8 51.f6 gxf6+ 52.Nxf6 Bxf6+ 53.Rxf6+ Kc5
Black decides to surrender his Bishop for the advanced passed pawn, submitting to the draw.
54.Rxa6 Rxe7+ 55.Kf4 Kb4 56.a4 Rb7 57.axb5 Rxb5 58.Kg3 Kxb3 59.Kh4 Kb4 60.Rh6 Rc5 61.Rxh5 Draw
Chessfriend Vlasta Fejfar of the Czech Republic has faced the "annoying defense" to the Jerome Gambit a number of times. In the following game, his most recent, he comes away with the whole point.
Vlastous - Idalgit
Internet, 2017
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
This is also called the "silicon defense" because it is the choice of many computer chess programs. Black returns a piece and takes a lot of the action out of the position.
8.fxe5 dxe5 9.Qh3+ Kf7
The text is about equal to 9...Ke7 which was seen in Fejfar,V -Goc,P, 2015 (1/2 - 1/2, 70), Fejfar,V - Chvojka, correspondence, 2016 (0-1, 32) and Vlastous - irinat, Chessmaniac, 2016 (0-1, 38).
10.Qh5+ Ke6 11.Qe2
Stockfish 8 gives the practical suggestion 11.Qh3+ hoping for a draw by repetition.
11...Qd6
Instead, 11...Ke7 was successful for Black in Wall,B - Alfil engine, Palm Bay, FL 2015 (0-1, 23); while 11...Nf6 was seen in Shredder 8 - RevvedUp, blitz 2 12, 2006 (1-0, 25) and RevvedUp - Yace Paderborn, blitz2 12, 2006 (0-1, 14).
12.Nc3
c6 13.Na4 Nf6
A tactical slip that drops a piece. Black may have unconsciously decided that his opponent has finished moving his Queen.
14.Qc4+ Ke7 15.Nxc5 b6 16.Nd3 a5 17.Qc3 Ke6
18.Qb3+ Black resigned
Perhaps a bit soon, but Black sees he will lose the b-pawn, and White's Queen will escape any danger, so the game may have lost its interest.
Once again, Vlastamil Fejfar, of the Czech Republic, plays the Jerome Gambit in a correspondence game, faces the "annoying" or "silicon" defense - and wins.
Fejfar, Vlastamil - Svoboda
corr Czech Cup, 2016
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
This "annoying defense" has been the choice of several of Vlasta's opponents in the past. See: Fejfar, V. - Chvojka, corr Czech Republic, 2016 (0-1, 32); Fejfar,V - Pressl, corr Czech Republic, 2015 (1/2-1/2, 15); Fejfar,V - Kyzlink, corr Czech Republic, 2015 (1-0, 22) and Fejfar,V - Goc,P, 2015, (1/2-1/2, 70).
8.fxe5 dxe5 9.Qh3+ Kd6
I pointed out before that at 25 ply, Stockfish 6 showed the tiniest preference for 9...Ke7 over 9...Kd6, but liked 9...Kf7 best of all.
To update this, I checked with Stockfish 7 at 30 ply, and found its preferences to be, in order, 9...Ke7, 9...Kd6 and 9...Kf7, but the spread was less than 1/10th of a pawn in evaluation.
For a more "human" insight, I checked The Database.
I found 47 games with 9...Ke7. Black scored 47%.
I found 20 games with 9...Kd6. Black scored 27%.
I found 75 games with 9...Kf7. Black scored 27%.
For practical purposes, and with the support of Stockfish, Black might prefer 9...Ke7.
10.Qd3+
10.Qg3 was Fejfar, V - Goc, Pavel, 2015 (1/2-1/2, 70)
10...Ke7 11.Qg3 Kf7 12.Qxe5 Bd7
This is an odd move. (Could it have been a typo for the more reasonable 12...Bd6?) It hands back a piece, giving White a two pawn advantage with little compensation.
Instead, ...Bd4 was seen in Fejfar,V - Goc,P, 2015, (1/2-1/2, 70), by a slightly different move order.
13.Qh5+ g6 14.Qxc5 Qh4+ 15.Qf2+ Qxf2+ 16.Kxf2 Nf6 17.d3 Rhf8 18.Nc3 Kg7
White needs to complete his development, but, otherwise, he has no problem - his King is relatively safe, his pawn structure is sound.
19.Ke2 Bg4+ 20.Ke3 Be6 21.h3 Nh5 22.Ne2 Rae8 23.b3 Nf6 24.Ba3 Rf7 25.c4 Rd8 26.Raf1 Rfd7 27.Nf4 Bg8 28.Bb2 Rf8
29.e5 Re7 30.Kd2 Ne8 31.e6+ Black resigned
Black's pieces trip over each other, and he is bound to lose a piece, i.e. 31...Kh6 32.Nd5!? Rxf1 33.Rxf1 Rxe6 34.Rf8 c6 (the Rook has nowhere to go) 35.Nf4 Re7 36.Rxg8
Nicely done.
I received another Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) game from chessfriend Vlastamil Fejfar, of the Czech Republic. It appears to be a correspondence game, and is a very difficult battle, ultimately being won by his opponent. It should be of interest to Readers.
Fejfar, Vlasta - Chvojka
correspondence (?) Czech Republic, 2016
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
The annoying "Annoying Defense", favorite of computer engines (hence the name, as well, "Silicon Defense"), which drains a lot of the play out of the opening and leaves White grasping at ways to continue his attack.
8.fxe5 dxe5 9.Qh3+ Ke7
For the alternative 9...Kd6 see Fejfar,V - Pressl, corr Czech Republic, 2015 (1/2-1/2, 15) and Fejfar,V - Kyzlink, corr Czech Republic, 2015 (1-0, 22).
10.Qg3 Kf7 11.Qxe5
11...Qh4+
Instead, 11...Bd4 was seen in Fejfar,V - Goc,P, 2015, (1/2-1/2, 70).
For 11...Qd6 see Wall,B - Qi,H, Chess.com, 2011 (1-0, 14); for 11...b6 see Wall,B - Redom,T, Chess.com, 2010, (1/2-1/2, 59).
12.g3 Qe7 13.Qxe7+ Nxe7
See "Philidor1792 vs The Annoying Defense" Parts 1 and 2.
White has two central pawns for his sacrificed piece, and would probably hold the game in club play or online blitz play. The fact that he can make no progress against his opponent in this game strongly suggests that it was a correspondence match.
14.h3 Be6 15.d3 Rhf8 16.Ke2 Nc6 17.c3 Kg8 18.Be3 Bd6 19.g4 Ne5
20.b3 Ng6 21.Nd2 Nf4+ 22.Bxf4 Bxf4 23.Nc4 a6 24.d4 Rae8 25.e5 b5 26.Ne3 c5
27.Rhf1 b4 28.Nf5 g5 29.Nd6 Re7 30.cxb4 cxd4 31.Kd3 Rb8 32.Ne4 Bf7 White resigned
Chessfriend Vlasta Fejfar sent a recent Jerome Gambit that he played. I am not sure if it is a correspondence game, like the previous three we have examined - see "Correspondence Play (Part 1)", "Correspondence Play (Part 2)" and "Correspondence Play (Part 3)" - but it is a long and difficult affair, showing the persistence and energy that sometimes is required of the gambiteer when his inital attack is warded off. His (expected) half-point is well-earned.
I have added some game references to assist the next Jerome Gambiteer who embarks on the journey.
Fejfar, Vlastimil - Goc, Pavel
2015
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+
4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ke6 7.f4 d6
Ah, yes, the "annoying defense". Black gives back a piece and snuffs out much of White's attacking play. Well, Vlasta has faced this before!
8.fxe5 dxe5 9.Qh3+ Ke7
Interesting. Fejfar,V - Pressl, corr Czech Republic, 2015 (1/2-1/2, 15) and Fejfar,V - Kyzlink, corr Czech Republic, 2015 (1-0, 22) continued 9...Kd6.
At 25 ply, Stockfish 6 shows the tiniest preference for 9...Ke7 over 9...Kd6, but likes 9...Kf7 best of all.
10.Qg3 Kf7 11.Qxe5
Also seen: 11.Rf1+ Nf6 12.Qxe5 Bb4 13.Qh5+ g6 14.Qf3 Re8 15.c3 Bf8 16.d4 Kg7 17.Bg5 Be7 18.e5 Ng4 19.Bxe7 Rxe7 20.h3 Nh6 21.Qf6+ Kg8 22.Qf4 Nf7 23.Nd2 c5 24.Ne4 cxd4 25.cxd4 Bf5 26.Nc3 Qb6 27.0-0-0 Rc8 28.g4 Bd7 29.Rd2 Qe6 30.Kb1 b5 31.Ne4 Kg7 32.Re2 Bc6 33.Nc5 drawn, Philidor1792 - NN, friendly match without time control, 2012.
11...Bd4
A number of games where Philidor1792 faced 11...Qh4+ can be found in the "Philidor1792 vs The Annoying Defense (Part 2)" post.
For 11...Qd6 see Wall,B - Qi,H, Chess.com, 2011 (1-0, 14); for 11...b6 see Wall,B - Redom,T, Chess.com, 2010, (1/2-1/2, 59).
12.Rf1+
An alternate is 12.Qh5+: 12...g6 13.Qf3+ Nf6 14.d3 Bg4 15.Qf4 Qd6 16.Qxd6 cxd6 17.Nd2 (17.c3 Bc5 18.Rf1 h6 19.d4 Bb6 20.h3 Be6 21.e5 dxe5 22.dxe5 Bf5 23.g4 Rae8 24.Kd1 Rxe5 25.gxf5 Rxf5 26.Rxf5 gxf5 27.Kc2 Kg6 28.Nd2 Re8 29.Nf3 Re4 30.Bd2 a6 31.Rf1 Bc5 32.b4 Ba7 33.Kd3 b5 34.a4 bxa4 35.c4 Nh5 36.Ra1 Nf4+ 37.Bxf4 Rxf4 38.Nd2 Bf2 39.Rxa4 Be1 40.Rxa6+ Kg5 41.c5 Bxd2 42.b5 Bb4 43.c6 Rh4 44.c7 Rxh3+ 45.Kd4 Bc3+ 46.Kd5 Rd3+ 47.Ke6 Re3+ 48.Kf7 Bf6 49.Rxf6 Black resigned, Wall,B - Harshini,A, chess-db, 2015) 17...Rac8 18.c3 Bb6 19.Rf1 Rhe8 20.h3 Bd7 21.Nc4 Rxc4 22.dxc4 Rxe4+ 23.Kd1 Rxc4 24.Bg5 Bd8 25.g4 h5 26.Bxf6 Bxf6 27.g5 Bf5 28.gxf6 Bxh3 29.Rf2 Bf5 30.Rd2 Rc6 31.Ke2 Kxf6 32.Ke3 Ke7 33.Re1 Rc5 34.Kf4+ Be6 35.Rde2 Rf5+ 36.Kg3 Rf6 37.Kh4 Kf7 38.Re4 a5 39.b3 Ke7 40.c4 b6 41.Kg5 Kf7 42.R1e2 Rf5+ 43.Kh6 Rf6 44.R2e3 Ke7 45.a3 Kf7 46.a4 Ke7 47.Rd4 Kd7 48.Kg7 Ke7 49.Rh4 g5 50.Rxh5 Rf7+ 51.Kg6 g4 52.Rh4 Rf6+ 53.Kg5 Rf7 54.Rh6 Rf5+ 55.Kxg4 Rf6+ 56.Kg5 Rf5+ 57.Kh4 Rf4+ 58.Kg3 Rf6 59.Rxf6 Kxf6 60.Kf4 Bd7 61.Ke4 Bc6+ 62.Kd4 Kf7 63.Rh3 Kg6 64.Rh8 Kf7 65.Rb8 b5 66.axb5 Bf3 67.b6 Ke6 68.b7 Kd7 69.Rd8+ Kxd8 70.b8Q+ Black resigned, Wall,B - Alibak, chess-db, 2015.
12...Nf6 13.Qh5+ Kg8 14.Qe2
Aha! A human being improves upon 14.Qg5 of Fritz 5.32 - Junior 7, The Jeroen Experience, 2003 (0-1, 32). Still, White has to be careful, with his King in the middle of the board, and with Black having much better piece development.
14...Be6 15.c3 Be5 16.g3 c5 17.d3 Qd6
18.Bf4 Bg4 19.Qe3 Bh3 20.Rf3 Bg4 21.Rf1
White is willing to repeat the position and agree on a draw.
21...Re8 22.Nd2 Bxf4 23.gxf4 b5
More worry for White. He holds on to the fact that Black's second Rook is still blocked out of play.
24.e5 Nd5 25.Qg3 Qg6 26.Ne4 c4 27.Kd2 Bf5 28.Nd6 Rd8
29.dxc4
Bold. I would have chosen the more solid 29.d4, but then Black would probably play 29...b4!?
29...bxc4 30.Rae1 Qxg3 31.hxg3 Ne7 32.Ke3 Bd3 33.Rg1 Nf5+ 34.Nxf5 Bxf5
The piece exchanges have reduced the danger to White's King (who is well-placed to support the "Jerome pawns") but Black still believes he can make something out of his piece vs 2 pawns advantage.
35.Rd1 Kf7 36.Rd4 h5 37.Rgd1 Rc8 38.Rh1 g6 39.Rh2 Ke7
The game has taken on an oddly closed nature. That does not last long, however.
40.a4 Rc6 41.Rd5 Be6 42.Rb5 Rc7 43.Rd2 h4 44.gxh4 Rxh4 45.Rd4 Rh3+ 46.Kf2 Rd3 47.a5 Rxd4 48.cxd4 Bd7 49.Rb8 Ke6 50.Ke3 Kd5
Black continues to push his small endgame advantage.
51.a6 c3 52.bxc3 Rxc3+ 53.Kf2 Bc8 54.Ra8 Rc7 55.Ke3 Rc3+ 56.Kf2 Kxd4 57.Rxa7 Ke4 58.Rg7 Kf5
Black does not want to trade off his last pawn (i.e. 58...Bxa6 59.Rxg6) but I think he misjudges the resulting reduced endgame.
59.a7 Ra3 60.Rf7+ Kg4 61.Rf8 Bb7 62.e6 Rxa7 63.e7 Bc6 64.e8Q Bxe8 65.Rxe8 Kxf4
66.Rf8+ Kg4 67.Rc8 g5 68.Rc3 Kh4 69.Kg1 Ra2 70.Rb3 g4
At this point the game was turned over to a referee for ajudication.
I am pretty sure that anyone familiar with Philidor's position will see that the game is drawn. Failing that, consulting the online Nalimov tablebases will also show that the point should be split.
Excelsior!
