he extra space
they take, involving more marble, more of the cutter's time and wear and
tear on his instruments, not to mention the big overhead, you would
think that Roman numerals would have been abolished long ago.

Of course, they can be figured out if you're good at that sort of
thing. By working on your cuff and backs of envelopes, you can translate
them in no time at all compared to the time taken by a cocoon to change
into a butterfly, for instance. All you have to do is remember that "M"
stands for either "_millium_," meaning thousand, or for "million." By
referring to the context you can tell which is more probable. If, for
example, it is a date, you can tell right away that it doesn't mean
"million," for there isn't any "million" in our dates. And there is
one-seventh or eighth of your number deciphered already. Then "C," of
course, stands for "_centum_," which you can translate by working
backwards at it, taking such a word as "century" or "per cent," and
looking up what they come from, and there you have it! By this time it
is hardly the middle of the afternoon, and all you have before you is a
combination of X's, I's and an L, the latter standing for "Elevated
Railway," and "Licorice," or, if you cross it with two little horizontal
lines, it stands for the English pound, which is equivalent to about
four dollars and eighty-odd cents in real money. Simple as sawing
through a log.

But it takes time. That's the big trouble with it. You can't do the
right thing by the office and go in for Roman numerals, too. And since
most of the people who pass such inscriptions are dependent on their
own earnings, why not cater to them a bit and let them in on the secret?

Probably the only reason that the people haven't risen up and demanded a
reform along these lines is because so few of them really give a hang
what the inscription says. If the American Antiquarian Turn-Verein
doesn't care about stating in understandable figures the date on which
the cornerstone of their building was laid, the average citizen is
perfectly willing to let the matter drop right there.

But it would never do to revert to Roman numerals in, say, the
arrangement of time-tables. How long would the commuter stand it if he
had to mumble to himself for twenty minutes and use up the margins of
his newspaper before he could figure out what was the next train after
the 5:18? Or this, over the telephone between wife and husband:

"Hello, dear! I