Oh, I think you mean where black's king is on d6, black plays e5 (to block a check by a bishop, and white plays d5xe6+ discovering both the Bishop on f4 check and the Queen on d1 check. Crazy.
The rarest move in chess ?

Unfortunately I can't insert a diagram from this computer for some reason, but a position in which the White King is on a5 and a White pawn had just advanced two squares to b4 while a Black pawns sat on a4, a Black Rook on a1 and a Black Bishop on e1 allows for axb3ep+ discovering checks from both the Black Rook and Bishop.

My rarest move was
Knight promotion check and he loses queen for free..
and also knight promotion mate
The raest move in chess has got to be the fried fox pork chop opening.
How can a named opening possibly be the rarest move in chess? And if that were the case, how is it not the Fool's mate? I've actually seen the fried fox pork chop played about a dozen games in a row (several of them against me).
I like that. 0-0-0# !!
I've done that in an OTB tournament game before.
You lucky, lucky b*****.

Well, we could go on arguing about this for hours...
How about Kxa8#... I don't think there will be many possible scenarios.
Or maybe Bc6xd5#, which would require three bishops, all of the same colour!

Ahh, yes, I suppose it could be interpreted that way -- I was thinking of a scenario where both checks are discoveries. Perhaps "double discovered check" describes this more accurately than "discovered double check".
The word discoved or discovery is probably redundent because there is no possible way to double check except through discovery anyway.

Ahh, yes, I suppose it could be interpreted that way -- I was thinking of a scenario where both checks are discoveries. Perhaps "double discovered check" describes this more accurately than "discovered double check".
The word discoved or discovery is probably redundent because there is no possible way to double check except through discovery anyway.
Well in this case both checks are discovered -- hence the use of double-discovered instead of discovered-double.
You mean the position? Bc6xd5# itself can happen at various less unusual positions. You can freely just remove c4 and e6 bishop.

You mean the position? Bc6xd5# itself can happen at various less unusual positions. You can freely just remove c4 and e6 bishop.
Incorrect. Then it would be Bxd5#. The other two bishops are required to add the c and the 6.
There will be several other positions (most with more pieces) but three white-squared bishops are required. (So two promotions to white-squared bishops must have occured.)

Ahh, yes, I suppose it could be interpreted that way -- I was thinking of a scenario where both checks are discoveries. Perhaps "double discovered check" describes this more accurately than "discovered double check".
The word discoved or discovery is probably redundent because there is no possible way to double check except through discovery anyway.
Well in this case both checks are discovered -- hence the use of double-discovered instead of discovered-double.
Oops. O.K yes I see what u mean now.
But that move might be common in games that have been recorded in long algebraic notation :P
But I see your point and nice idea.

I think we are assuming short-algebraic notation! i.e. the notation used on this site. So Bc6xd5# is exceptionally rare.
For some of the stuff here, take a look at these pages (from Tim Krabbe's amazing website):
http://www.xs4all.nl/~timkr/chess2/minor.htm for a bunch of underpromotions in serious games. He shows about a dozen non-trivial examples. The final example, underpromotion to Bishop, is truly astonishing and a very deep move.
For very rare moves, look at entry 105 at http://www.xs4all.nl/~timkr/chess2/diary_6.htm .
In the case of en passant, how is the pawn capture a discovery?