i.                    P-K4           P-K4

ii.                  Kt-KB3       P-KB4

iii.                KtxP           B-QB4

iv.                 Q-KR5 ch.  P-KKt3  *

v.                   KtxP           Kt-KB3

vi.                 QxPKB5

*(iv. …K-KB1 v. Q-KB7 mate, iv. …K-K2 v. Q-KKt5 ch. K-K1 vi. QxQ ch. KxQ Kt-KB7 ch. 1-0)

In this position (arrived at by several moves that are absolutely forced for black), taking the knight and then letting the queen take your only attacking piece looks like a mistake, so I tried:

vi.                                                                               …P-Q4

vii.                  Q-K5 ch.                K-KB2

viii.                KtxR                       QxKt

ix.                 QxPQB7 ch.           Kt-Q2  

Where after PxP KtxP the queen will only have two safe places and will be the only white piece developed. But white will be ahead the equivalent of two minor pieces! Finding compensation for black in this position was hard. Eventually I reached this position, however, which shows just how well the pieces can cooperate together. (Unfortunately, although I've come close, I couldn't recreate the moves creating the position, since I didn't write them down at the time and was counting on being able to recreate it from scratch--a bad habit). However, I'm not sure white can't avoid this position; here it is anyway:

(Position 1) 

Here it seems that Black has the game won. The white queen has no safe spot to check on, and if white stops the checkmate threat by R-K1, then Kt-KB7 is checkmate. Unfortunately, even in this position, White has a tactical shot:

i.                     Q-K6 ch. !!   K-KKt4  *             **           ***

ii.                   QxB ch. !!    QxQ       ****

iii.                  R-KKt1

*             Where the king goes is not really important

**           i. …BxQ? ii. Kt-KB4 ch. 1-0

***        i. …Kt-KB3 ii. R-K1

****     ii. …KxQ doesn’t change matters

So, spectacularly, this variation seems to have been 'cooked' (although I will admit your opponent might not find the last move that I outlined). But I knew there had to be a way. iii. ...B-QB4 had to be the right move. It was too awesome not to be. So I then carefully went over the analysis and realised that I'd missed a move that immediately equalises (at least). After:

i.                    P-K4           P-K4

ii.                  Kt-KB3       P-KB4

iii.                KtxP           B-QB4

iv.                 Q-KR5 ch.  P-KKt3  *

v.                   KtxP           Kt-KB3

vi.                 QxPKB5

vi. ...P-Q4?! is probably a mistake, as previously outlined. Better is vi. ...P-Q3! which covers the checking square. Hence the following game should ensue:

(Game 2)

So it turns out that not only is the Latvian Gambit upheld, rising like a phoenix from the ashes, but that we have a definitive answer (one with many forced moves, allbeit) to the queen check on the third move. If only there can be an answer for the bishop move, then you could practically force your opponent to take the second pawn and initiate the bishop sacrifice like I outlined in 'Latvian Revenge'. But this line should definitely help Latvian Gambit players!