Both "ghost" numbers and other embossing (e.g. names of companies) are found on
several soda bottles I have worked with, and I suspect George's interpretation of why
they happen is correct.  In the early part of the 20th century, soda bottlers often had
their bottles embossed with either the company name or theinitials of the owner.
These often show up in "ghost" form as well.

Although unrelated, another base phenomenon is interesting.  On the bases of milk
bottles, valve or ejection marks often appear.  These are small, distinct circles near
the center of the base.  They often contain crisp numerals in reverse (although they
can also be indestinct).  Although I have not looked into these seriously, I see them
mostly in milk bottles from the 1930s and 1940s.

There are numerous other possibly datable aspects of bottle embossing that still
need to be researched, especially with late 19th and all 20th century bottles.

Bill Lockhart

> Chris Murphy's comment on "ghost" mold numbers in the bottom of jars
> rings true.  These "ghosts lines" are commonly found on canning jars
> that are clearly mouth-blown.  My guess is that the parison of glass
> on the blow pipe  hit the bottom of the mold before the bottle was
> completely blown. Then when the bottle or jar is fully blown, the base
> shifts as the air pressure pushes the glass against the walls of the
> mold.  These ghost mold numbers are from the blow mold which is
> different from those produced by a parison mold.
>
>       With machine-made bottles, both semi-automatic and fully
>       automatic
> there is a parison mold that forms the glass around the blow pipe.
> That mold has seams, which become compressed when the parison is moved
> to the full or blow mold.  In many cases, the parison mold lines are
> in alignment with the mold seams of the blow mold, so that the parison
> mold can not be distinguished from the blow mold.  The easiest place
> to see the parison mold is in the base of the vessel.  It usually is a
> circle that is slightly off center of the base.  The Owens scar is
> from the parison mold and in a sense is a ghost mold line.  In some
> cases, when the knife that cuts the glass off during the gathering
> process is working well and the glass is just the right temperature,
> the base may not have an Owens scar.  I have cataloged two identical
> bottles with the same mold numbers and shape where one has a visible
> Owens scar and the other does not.  Thus, we can be sure that a bottle
> was made on an Owens machine if the scar can be seen, but there are
> some bottles that were made by that process where the scar will not be
> readily apparent.
>
>       Tony McNichol and I have been expanding our paper on the
>       chronology
> of early machine-made bottles and would be interested in knowing about
> any published pamphlets and price list data from Owens Bottle Company,
> Illinois Glass Company, and Owens-Illinois Glass Company.   We are
> also interested in journals from the glass industry.  Any help with
> this would be greatly appreciated.
>
> George L. Miller
> URS Corporation
> 561 Cedar Lane
> Florence, New Jersey  08518

Bill Lockhart
New Mexico State University
Alamogordo, NM
(505) 439-3732