ErictheRed wrote on 03/21/15 at 18:45:58:


MCO mentioned Larry Evans using it in the 1950s and '60s; I think my first awareness of it may have come from his books (I know he used part of his game against Bronstein in New Ideas in Chess).  Don't know of anything special, alas.

ErictheRed wrote on 03/20/15 at 20:43:27:


Ah, the Romih variation, as I recall it being called in books from some decades ago.  It reminds me of some stuff that's been happening in the 4. e3 Nimzo.

TN wrote on 03/19/15 at 20:12:55:

This is not entirely correct. 8.Bd3 c5 9.a4 b4 10.Ne4 Nbd7 and 9.O-O Nbd7 10.e4 (hoping for Bb7 11.d5, Reynolds Attack) cxd4 transpose back to the Meran. These sidelines are pretty good for Black though.

The basic idea is that you are making a semi-useful waiting move to see what White's setup is before committing yourself - so you can meet 6.b3 with 6...Bb4 to provoke the White bishop to the bad d2 square, and after 6.Qc2 you can play 6...c5 to open the c-file where White's queen will be misplaced (especially if an IQP position arises). If White plays 6.c5, the pressure on Black's centre is released and therefore you can undermine White's structure with either 6...b6 or 6...Nbd7 intending ...e5. Finally, you avoid the Meran as after 6.Bd3 dxc4 7.Bxc4 b5 and 8...c5 you end up in a QGA with White committed to an early Nc3, which isn't anything special for the first player. 

Scherbakov has covered this 5...a6 move quite thoroughly on ChessPublishing, but there have been enough recent games to possibly justify a brief update in the near future. Andreikin is the main exponent for Black of late, which isn't surprising given that the main aim of this variation is to gradually equalise.