You rarely want to have a variable name itself be variable, for many good reasons. If you don't know how to do it, you probably don't really want it. (I'll insert a link to a node that describes the technique and caveats as soon as I [id://Super Search|find it] again. Also, I'm not being snotty, I don't know how to do it either.)

However, the technique described by madbombX and mreece is usually the best way to do the job. Here's a quickie example on how to use the technique:

#!/usr/bin/perl -w
use strict;
use warnings;

my %hr = (
    'default' => [ 1 .. 9],
    'even'    => [ 0, 2, 4, 6, 8],
    'odd'     => [ 1, 3, 5, 7, 9],
);

my $arrname = shift || 'default';

print "Array $arrname contains:\n";

my $ar = $hr{$arrname};
for my $cnt (0 .. $#{$ar}) {
    printf "\t %s[%d]=%s\n", $arrname, $cnt, $ar->[$cnt];
}
[download]

Running this yields:

roboticus@swill ~/PerlMonks
$ ./var_array_names.pl  even
Array even contains:
         even[0]=0
         even[1]=2
         even[2]=4
         even[3]=6
         even[4]=8

roboticus@swill ~/PerlMonks
$ ./var_array_names.pl  odd
Array odd contains:
         odd[0]=1
         odd[1]=3
         odd[2]=5
         odd[3]=7
         odd[4]=9
[download]

--roboticus

UPDATE: Yarch! I was a bit slow, and several people beat me to the punch. Oh, well. Anyway, here are links to a few of the threads describing the pro & cons: