RE: %Elem(Array) vs. cCMLimit

Constants are always better than hard-coding the value, so 
that's a great job in and of itself.
Using %ELEM() in your code is more clear, only because 
cCMLimit is an ambiguous name, as re fields with special 
symbols in them, such as CM# and IN#.

It was the unambiguity of %Elem(ArrayName) that suddenly popped into my head
today for who knows what reason.  One of those Aha! things that you wonder
why you never thought of it years ago.

Certainly %elem(CM#) is clear that you're using the number of 
declared elements for the array CM#, but what does that array 
contain? 

CM# is Credit Memo Number.  Even though I converted everything to RPGIV with
a conversion tool eons ago, some field and array names are still remnants of
fixed-form RPGIII coding where stuff like "CM#,10" had to fit in 6 columns.
I do not miss those days. <g>  I can't hardly stand looking at fixed-format
RPGIV any more, let alone RPGIII.
 

Another approach (not a better one, but one of equal honor) 
is to use %ELEM() everywhere instead of a const. For example:

D cCMLimit        C                   Const(20)
D CM#             S              7  0 Dim(cCMLimit)
D IN#             S              7  0 Dim(%elem(cm#))
D AMT             S              7  2 Dim(%elem(cm#))

That's the one I like.

Then in your code also use %elem(cm#). 
The benefit of this method, is that it directly says:
"Arrays IN# and AMT have the same number of elements as CM# 
and this was done on purpose and is a requirement." 

And that's why it's the one I like.

Thanks to all for everyone's 2 cents worth.