Concentration units: g/L, mg/ml, ug/ul. ng/nl, or what?

Forgive the off-topic post, but I thought this audience might
be tolerant of my complaint- if you disagree, delete me at once.

I just got the galley proofs for a book article, and discovered
that the copy-editor has replaced everywhere my concentration
units of g/L with mg/ml. Granted the volumes were not in most
cases liters, but they weren't ml either (more like ul).
And the whole idea of concentration is the ratio of mass to
volume, in order to give a volume-independent description of
the chemistry. "mg/ml" looks like someone forgot to cancel the
m's when they divided. In grade school algebra when I handed
in a problem with a result like "1000 X = 5000" I would get
points taken off for not canceling the 0's.  Do we really have
to express concentration in different units depending whether
we work with liters, ml, ul or nl? The SI system has a single
concentration unit "molar" which can and must be used whether
you work with liters or nl.  This concept of volume-independent
concentration is not so difficult, it's sort of like 
distance-independent speed. Do you think I could get a judge 
to dismiss my ticket for driving 50 miles per hour on elm street 

a: elm street is only 10 miles long,
b: I had only been driving for 10 minutes when I was stopped,

and so obviously I couldn't have been going 50 miles per
hour on elm street?

Do we need to have a second speedometer calibrated in m/sec
for driving on roads less than one mile (km) long or for
time periods less than one hour?

Sorry for the outburst about a trivial peeve, at a time when
there are more serious things to be concerned about, but I
would be interested what others think. One obvious response
would be: no, we only need one unit of concentration, but mg/ml
is preferred over g/L because of the euphonic way "migs/em-el"
rolls off the tongue or because the proper SI abbreviation for
liter is a lower case script "L" which is not available in usual
fonts, and lower case L looks too much like the digit "1",
but ml is unmistakable. And I think I saw once, in a very old
instructions to authors, an admonition against using "deceptive"
terms like grams/liter when the actual volume is much less.
The implication is that I am "deceiving" my readers by pretending
I can make grams of this protein when I never saw more than
100 ug in my life.


Oh, and another unit I don't approve of is "% weight/vol".
The only good thing about percent is that it gives you a 
clear idea of the fraction of the total material comprised 
by the component. In particular, if you add up all the
percentages, you should get 100 if everything is
 accounted for. And if one component comprises 100%, 
 it should be pure, "neat" stuff.  Neither of these 
 is true of %w/v. Neat Glycerol is 126% w/v. MeI is 228% w/v.
 And 100% w/v TCA, which you can buy from Sigma, is a liquid 
 (unlike the neat stuff which is crystalline solid).

And % v/v is only a little better, because volumes aren't 
always additive. There is the question whether 50% v/v
of A in B means equal volumes of A and B were mixed, 
or whether A was diluted with B until the final volume was
twice the starting volume of A.

Better to use g/L instead of % w/v, and ml/L instead of % v/v.
This means that if you have one L of the soution, it contains 
that many grams, or that many ml, of the pure substance.