So far I've got around to covering all lines except for 3...Nf6, but I expect to follow up with three practical examples of 3...Nf6 in the very near future. The link is here:
http://tws27.weebly.com/morra-gambit-declined.html
Unlike most other sources on the declined variations of the Morra, I have devoted rather more coverage to 3...d5 than to 3...d3, a decision which was partly influenced by some analysis by Mark Nieuweboer illustrating that it is harder for White to claim a substantial advantage than many sources (including Esserman) suggest. I don't think much of 3...d3, which allows White to set up a favourable version of the Maroczy Bind, such as in the following diagram:
3...d5 probably does not equalise fully, since by comparison with the Danish Gambit Declined, Black is one move further away from developing the kingside pieces (due to having played ...c7-c5 rather than ...e7-e5), but I think that with accurate play White only gets a small edge in a typical "isolated queen's pawn" situation, with active piece play providing compensation for having an isolated pawn on d4. After 4.exd5 Qxd5 5.cxd4 Black has to choose between setups based on ...Bg4, ...e6, or ...e5. I think the continuation with 5...Nc6 6.Nf3 Bg4 is a bit suspect due to the sharp ending that arises from 7.Nc3 Bxf3 (7...Qa5 8.d5) 8.Nxd5 Bxd1 9.Nxc7+ being favourable for White, but after either 6...e6 or 6...e5 (the immediate 5...e5 is also worth considering) Black gets quite a reasonable game, though probably falls short of full equality.
I have not yet examined the 3...Nf6 lines (which, after 4.e5 Nd5, transpose directly into the Alapin variation of the Sicilian, 2.c3 Nf6 3.e5 Nd5 4.d4 exd4). I have found three attractive attacking high-level games from White's perspective but my preliminary impression is that theoretically Black should be doing fine.
A little recap on the 3...dxc3 lines
Daniel King was quite dismissive of the gambit in his recent "How Good is your Chess?" article for Chess Monthly, referring to the line 1.e4 c5 2.d4 cxd4 3.c3 dxc3 4.Nxc3 Nc6 5.Nf3 d6 6.Bc4 a6 7.Bg5 Nf6 8.Bxf6 gxf6 9.0-0 Bg7 10.Nd4.
I didn't even mention the line starting with 7.Bg5 in my own coverage, but I have no regrets about this, since as King points out, this position is an Open Sicilian position (1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 Nc6 7.Bxf6 gxf6 8.Bc4 Bg7 9.0-0) but where White is missing a pawn on c2, and thus it is no surprise that White struggles to obtain decent compensation for the pawn. 7.0-0, intending 7...e6 8.Bf4, is the way to go against the Taylor Defence starting with 6...a6.
However, my opinion of the Morra has dropped a little after, with the help of Mark Nieuweboer, uncovering a few move-order tricks for Black which, while not really changing the assessment of the line from "dynamically equal", suggest that there may be certain lines where the onus is more on White to prove full compensation for the pawn, than on Black to prove equality. Nonetheless, I expect to continue employing the Morra at times, alongside the Open Sicilian, for the foreseeable future.
