I think 5...c5 is not consistent with this opening, laying the foundation for all of Geller's problems, and 5...d6 would be more appropriate (keeping the game closed in lieu of Keres' control of the center).

Usually in the KI one would play 4...d6 and play would progress from there. After 5.f4 (KI, 4 pawns) the usual play is 5...0-0 6.Nf3 and then 6...c5, a transposed move order as seen in the Geller-Wade game. For example, Bisguier vs Fischer, 1956

The only other games in a search of 5...c5 yields some pretty ugly games, in favor of white as I predicted. The only real games of note for black are 2 games in which white does NOT play 6.e5

I dont think Fischer's win necessarily means that white never played 6.e5 thereafter, or that 6.e5 was refuted after 1960. In the Wade-Geller game for example, Geller is not inviting 6.e5, in fact he plays 5...d6 to prevent 6.e5, and the game was played in 1963.

According to your point, Geller should have drawn white in by inviting 6.e5 if it meant winning lines for black.

Which brings us full circle to the game at hand! In the above game Geller plays 5...c5, white plays 6.e5, and Keres <CRUSHES> poor Dr. Geller!

The tactics at the end of Letelier-Fischer are cute, but its real significance (it was hugely influential at the time and it appears in M60MG), is not only to show the right strategy for Black in these sorts of positions (ie dismantle the centre rather than play around it) but to generate a shift in what was judged to be a strong and what an over-extended centre. There were some other games around the same time to the same affect: Barden Hartston Keene on the KID (2nd ed page 1973 243-4) give Boznipa-Stein, Harkov 1959 and Elkon-Vaskan Riga 1960 but I don’t have them to hand – does anyone?)

Of course I wasn't entirely serious when I said white never subsequently played 6.e5. Better the line lost popularity.

Yes, they would be identical. I had to set up both games, then play the Geller game with different moves to see it, but you are right, the Geller game could transpose into the Fischer game. Thanks for pointing that out.

Also, I believe that you are more correct; 7...Qa5 is more of a losing move than 5...c5, since 5...c5 does not in and of itself lead to the loss, and with proper play, black can salvage the position (as long as you are Bobby Fischer in 1960). Since 7...Qa5 is not possible, though, without 5...c5, I will say instead that Geller misplayed the opening. I originally thought that 5...c5 is not consistent with this opening (which I still think) but more correct is that 7...Qa5 is not consistent with 5...c5: "the blunder of a bad opening"

I still feel that since the KI is primarily a 4...d6 system, 5...c5 is not a wise choice in the KI system, 5...d6 is a better choice, and has been outplayed 622 to 12 in the database in the above position at black's move 5. (KI with f4). (The KI with f4, for that matter, is not the main line for white, either.)

I still feel 6.e5 is good for white and would play 6.e5 whenever given the <choice>, again, assuming that black gives me the choice by not playing 4...d6 or 5...d6. In other words, in the above position, I would intend to play 6.e5 in a heartbeat as white.

Fischer unhinged Letelier's play masterfully, but to me, this still does not mean that black has anything but big problems after 6.e5. I would never play a move like 5...c5 as black, and expect anything less than 6.e5, then think that I could extricate myself from it as Fischer did.