# Re: Arrow heights

• To: math-font-discuss@cogs.susx.ac.uk
• Subject: Re: Arrow heights
• From: alanje@cogs.susx.ac.uk (Alan Jeffrey)
• Date: Mon, 23 Aug 93 02:04 BST

Barbara writes...

>knowing that knuth spent a lot of thought on how these symbols
>should behave, and also knowing that not all of the criteria that
>he considered were spelled out in the mf code,

Yes, the CM arrows are quite well-behaved.  There is still a noticably
large gap between the arrow and its labels in:
$\mathop\rightarrow\limit^x_x$
If you try the same trick with \Rightarrow you get much better
results, because the height of \Rightarrow is at the top horizontal
line, rather than the top of the arrow.

I've been thinking about why DEK didn't do the same with \rightarrow,
and I'm guessing that he wanted \rightarrow and \Rightarrow to be
visually similar when decorated with scripts, so:
$\rightarrow^x \Rightarrow^x$
produces superscripts at the same height.  Any better ideas anyone?

I've also noticed that \rightarrow and \Rightarrow suffer rather badly
if put in a \vcenter, since the centre of their box isn't on the math
axis.  Is there a reason why \rightarrow and \Rightarrow weren't given
negative depths, so they would \vcenter properly?

Alan.