In Autumn 2013 edition of Patent issues, we wrote about how a priority document can “poison” a later application claiming priority from it. This can happen where the two applications cover the same territory and the claim to priority is invalid under the strict “same invention” rule because the earlier application is prior art to the later one under section 2(3) of the UK Patent Act 1977 or Article 53(4) of the EPC.
And if a priority document can poison a later application, so too can a divisional poison a parent (or vice-versa) because parent and divisional are considered separate applications.
We also wrote about the need have claims that clearly demark (using the word “or” if necessary), those parts of the claim that are entitled to priority and those that may not be, so that any parts that are not entitled to priority are not anticipated by the priority document.
We are pleased to report that these issues have been referred to the Enlarged Board of Appeal of the EPO in appeal G 1/15.
In particular, the Enlarged Board is asked to review its earlier decision in G2/98 whereby partial priority is only recognized where a claim encompasses a “limited number of clearly defined alternatives” (some being entitled to priority and others not). The problem arises where the species that is entitled to priority is later claimed (or separately claimed in a divisional application) within a broader genus. Some EPO decisions allow such a claim to be conceptually separated into the species that is entitled to priority and others that are not. But other decisions are less generous, and it is this divergence that the EBoA is to review.
This review is welcomed, and its outcome is keenly anticipated. The Appellant, Infinium USA, wrote “a decision of the Enlarged Board on partial priority is of fundamental importance considering the potential impact . . . on applicants’ filing strategies”. Infinium argues that a “practice of self-collision of parent and divisional applications” is inconsistent with public policy considerations. We very much agree.
Our advice in the meantime remains the same:
• generic expressions will not be supported by specific examples;
• it is better not to rely on the G2/98 test for awarding multiple priorities for generic expressions that comprise a limited number of clearly defined alternatives;
• call out the specific alternatives so that each has its respective priority date.