Spontaneously, I came up with a cheating idea :
Instead of capturing the number, you could try to throw away the name of the chemical :

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

while (<DATA>) {
  chomp;
  if (/^(.*?)\s[A-Za-z]+$/) {
    print "Found $1 in $_\n";
  } else {
    warn "Don't know how to handle $_\n";
  };
};
[download]

This regular expression is of course very crude, as it will accept anything before the first whitespace as the number and discard everything after that. If your data dosen't allow this "easy-way-out-solution, you could try to gather all of your different regular expressions into one big regular expression by putting them into non-capturing parentheses and concatenating them with alternation :

  if (m!^((?:\d+)|(?:\d+/\d+)|(\N\+1\))|\d*[MN])\s!) {
    print "Found $1 in $_";
  };
[download]

If the above big RE works for you, you can start optimizing it and collecting same parts together, for example the first part \d+ is just a sub-part of \d+/\d+ and the two could be folded into one using the optional operator ? like :

m!\d+(/\d+)?! # Here, we use capturing parentheses for legibility
m!\d+(?:/\d+)! # Noncapturing parentheses for programmer-ease
[download]

If you want more information, the perlre manpage has more information, but if you think you really want to use regular expressions as a frequent tool, there is no way around Jeffrey Friedels Mastering Regular Expressions (but it's rumored that there will be a new version in the near future...)

perl -MHTTP::Daemon -MHTTP::Response -MLWP::Simple -e ' ;    # The  
$d = new HTTP::Daemon and fork and getprint $d->url and exit;#spider
($c = $d->accept())->get_request(); $c->send_response( new   #in the
HTTP::Response(200,$_,$_,qq(Just another Perl hacker\n))); ' #  web
[download]