[h-e-w] Displaying Unicode - font substitution for missing glyphs

[John Brown]

Hello All,

I tested emacs 24.2.1 on Windows (8) with

http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-demo.txt,
which, as the name suggests, is a UTF-8 encoded file that
contains text in various languages, and special symbols such as
the euro.

I found that Courier New contained several missing glyphs, so
I experimented with other fixed-width fonts and got fewest
unrecognised characters with DejaVu Sans Mono.

Firefox displays all the characters, even though its default
fonts "Times New Roman", "Arial" and "Courier New" did not 

do well in emacs.

I found
http://lists.gnu.org/archive/html/emacs-orgmode/2011-07/msg00624.html,
where the poster said:

"I /think/ that in X, emacs will select the
closest font it can find to in order to get the characters it needs.
However, in Windows, it will only use the default font (or whatever
is explicitly specified for the face), even if that font is missing
characters."


Another poster in that thread said in
http://lists.gnu.org/archive/html/emacs-orgmode/2011-07/msg00646.html
that font substitution seemed to be working for him with a
binary that he got from 

http://code.google.com/p/emacs-for-windows/ 


On Linux (Ubuntu 12.04 - Precise Pangolin) I see that emacs 23.3.1 

displays all the symbols in the default font (whatever it was; I
did not have to change it) and in any other font that I tried,
although on Linux only a handful of fonts were available.

Why is emacs able to display all the characters on Linux but not
on Windows?

Regards,
John Brown.