Allen

I'm not sure it's accurate to list singletons as "strays" just
yet. Let's wait until we have all 100 data sets before we can
get a full picture of the reporting marks distribution. To get
accurate confirmation of Tim Gilbert's thesis of proportional
distribution of box cars, we'd really need to know which of the
reporting marks belong to box cars. By his thesis the number of
off-line box cars should be close to the percentages of those
reporting marks in the national fleet.

ALTHOUGH: The data may be quite skewed by the postwar recession!
Just as during a depression, recessions often result in off-line
cars being sent back home much sooner, and empty, instead of
being loaded over and over again off-line for new destinations.

IOW it may be fun to sift all this data but unless one models
this part of the country in 1947, it may have no other use.

Tim O'Connor