Re: German patent numbering system

Stephen Adams (stevea@magister.co.uk)
Fri, 12 Feb 1999 19:17:12 +0000

It looks as if this is a fairly spectacular case of mis-citation - as
others have pointed out, under the current system, the number at first
publication stage (DE-A1) is retained in Germany through to grant
(DE-C2), and it's also been the same as the application number since the
deferred examination system came in in 1968.

The historical sequence is as follows :
Up to 1957 - one publication stage only, Patentschrift
1957-1968 - first publication for opposition as Auslegeschrift, followed
by Patentschrift with same number if not opposed.
1968-1981 - published unexamined as Offenlegungschrift (DE-A), followed
after examination by Auslegeschrift with same number (DE-B), followed by
Patentschrift with same number (DE-C) if not opposed.
1981 onwards - Auslegeschrift stage dropped ; most published unexamined
as Offenlegungschrift (A1) followed by Patentschrift with same number
(C2). Some go straight through to Patentschrift (C1).

German numbering has several discontinuities in it - it rises from no.1
in 1877 to around the 768,000 level by 1945, at which point granting
stopped for several years. The next grants were from 800,001 upwards,
dating from 1950. When deferred examination came in in 1968, the range
jumped again from around 1,200,000 to 1800001. At this point, it ceased
to be a continuous series, and comprised a 2-digit year prefix + 5 digit
annual series ; simply add 50 to the first 2 digits to get the year of
application.

I know it's too late now, but I have always found it useful to have a
concordance table handy to check whether apparent "equivalents" were at
least published in the same time-frame (not infallible, but can be
quicker than ordering the wrong document). If you are dealing with
cited art rather than equivalents, of course, this argument breaks down.

I have an old table (produced by Chas. Hude, the Danish patent attorney
practice but sadly long discontinued) which shows up that US 3941932
published some time after 1968 (off the bottom of the table) whilst DE
967167 was in the 1956/57 range. Unless the US case was drastically
delayed in prosecution, I'd start to smell a rat at that point.

As regards databases, EDOC on Questel covers DE back to 1877 [sic], and
although you can't do a family check until 1969+ publications, the
bibliographic details of DE 967167 should include a classification mark
which you could cross-check in ECLATX to see if it's the right technical
field. Unfortunately, EDOC does not include inventors so you could not
search for the genuine number in this case.

--
Stephen Adams
Magister Ltd. - Patents Documentation Consultancy
62 Norton Road, Reading, RG1 3QJ, GB
Tel/fax:  +44 (0)118 926 7981
e-mail:  stevea@magister.co.uk


Registered in England and Wales.  Company No. 3407685
Registered address : Canada House, 272 Field End Road, Eastcote,
Ruislip, Middlesex  HA4 9NA.