Is there any difference between prototype % and @ ?

LanX has asked for the wisdom of the Perl Monks concerning the following question:

in Re: Is there a simple syntax to logically slice a hash? I used a '%' in the prototype of a grep like function

sub hgrep (&%)
[download]

I somehow expected it to better than sub hgrep (&@) by warning if the number of passed list elements is odd.

But neither from tests nor documentation can I find any differences...

Is it just cosmetics?

Cheers Rolf

UPDATE: of course I'm not talking about the documented behavior for \% or \[%]