Added matter, intermediate generalisation and a good day for cricket

After a tennis-related post earlier this morning, it is tempting to link this article to another of England's frequently rain-affected summer sports, cricket. Sadly, this doesn't quite work. The Grace in this case is not the legendary WG Grace (right) by the presumably unrelated WR Grace -- it's a patent-y sort of case with lots of facts and not a huge amount of intelligible law. The case in question, decided by Mr Justice Floyd yesterday in the Patents Court, England and Wales, is Fosroc International Ltd v WR Grace & Co.-Conn [2010] EWHC 1702 (Ch), and it's largely about added subject matter.

This post isn't even going to try to explain the invention and the details of the patent in its pristine and putatively amended state. It was something to do with concrete (or was it cement?) -- or at any rate it was to do with substances that Kats don't like putting their paws into when it''s wet. The court essentially held that the patent was valid and that the amendments would be allowed. In doing so, Floyd J summarised the main principles involved when considering whether an attempt to amend a patent would fall foul of the "no added matter" rule. In short:

* When considering whether a proposed amendment added any matter to the patent, the court had to see first what was disclosed in the patent and then look at what was disclosed in the patent as amended. It should then compare the two disclosures and decide whether any subject matter relevant to the invention had been added [No rocket science here, in principle at any rate: it looks like a simple subtraction exercise].

* If the specification discloses distinct subclasses of the overall inventive concept, it should be possible to amend by whittling the patent down to one or more of those subclasses, regardless of whether they were presented as inventively distinct in the specification before amendment.

* when a proposed amendment seeks to take features which are only disclosed in one particular context and which are not disclosed as having any inventive significance, introducing them into a claim which is not limited by that context, we have "intermediate generalisation", a subtle, sneaky form of extended subject matter which has to be watched out for.

The IPKat has never been terribly sure what "intermediate generalisation" means, even if it gets the occasional mention on this weblog (see eg here), so he's hugely grateful for the clarification.

Cat's paw cement here