perldoc perldata has a good section on hashes. Think of a hash like a dictionary or a phone book. You look up a word or a name, and you get back a definition or a phone number.

With a hash, you use a string as the key (what you put in) and you get back a scalar as the value. In this case, the keys are filenames and the values are filehandles. There's a one-to-one mapping of keys to values.