It sends them a very nice rejection letter, as Heinlein notes in his 1958 Have Space Suit---Will Travel (his juveniles were intended to teach readers what they

It sends them a very nice rejection letter, as Heinlein notes in his 1958 Have Space Suit---Will Travel (his juveniles were intended to teach readers what they needed to know to go out into space, from attitudes to what they needed to learn. The latter was touched upon in the first, the 1947 Rocket Ship Galileo, this book gave it a very thorough treatment in the beginning when the (genius, scientist, etc.) father of the protagonist realized how awful was the public school education he was getting). And as I can attest, there's this nice bit about how they have many fewer spaces than qualified applicants, not going into the minor detail of if you were one of the latter (I wasn't). CalTech is very special, even more so than MIT I gather (I'm Class of '83 of the latter). Much smaller, 1/4 the class size (~250 vs. 1100), much more intense, much more science focused. Someone, commenting on how they'll clean up your room every week (a service MIT is sadly lacking in :-), said it was like a Hogwarts for science, and very simply, if you can't do magic at its level, they don't accept you. For some historical perspective, as I understand it, at the post-Civil War beginning of MIT, most students were mastering the calculus by the end of their undergraduate program. (And our "freaky" unit numbering is great, totally transparent, and at least at MIT there's a committee dedicated to the task of keeping professors and departments in line with the Institute's limits on work demanded, and has been known to take classes away from abusive professors. That's also done at the departmental level, as I witnessed one semester while on the EECS department's staff (took over and finished the sysadmin part of the job of moving it from Multics to UNIX(TM), with the help of one on and off student like me who was an old friend).)