The European Patent Office (EPO) recently issued its reasoned decision clarifying the EPO’s stance regarding partial priority (sometimes called multiple priorities).[1]  But what is the relationship between poisonous priority and partial priority and how do the legal approaches to these issues vary between different jurisdictions?  We take a look at these aspects of patent law in the context of the European Patent Convention (EPC) and the laws in Australia and New Zealand.

Prior rights and poisonous priority

Under the “prior rights” principle, if patent application A has a filing date which is before the filing date of application B (in the same jurisdiction) but A has a publication date which is after the filing date of B, the whole contents of A form part of the prior art for the purposes of assessing novelty (but not inventive step) of B’s claims.[2]  This principle gives rise to “novelty only” prior art[3] (which may alternatively be referred to as “whole of contents” prior art).[4]

This principle not only applies to unrelated prior art, but also has the potential to cause one patent application to invalidate claims of a related application.

In a simple scenario, the theory of so-called “poisonous” or “toxic” priority suggests that self-collision of a parent application with its divisional child application might arise as follows:

  • A first application, X, is filed, disclosing a specific widget.
  • A second application, Y, is filed and claims priority from X. Y discloses the same specific widget and, for the first time, a general embodiment of the widget.  Y claims the general embodiment.
  • A divisional application, Z, is filed having Y as its parent, claiming priority from X and containing the entire disclosure of Y. Divisional application Z is published.
  • According to the poisonous priority argument, the disclosure of the specific widget in Z (which is entitled to priority, having first been disclosed on filing of X) is “novelty only” prior art[5] against Y’s claim to the general embodiment (the general embodiment having first been disclosed later, on Y’s filing date). Thus the disclosure of the specific widget in Z destroys the novelty of Y’s claim to the general embodiment.

For Y’s claim to the general embodiment, whether or not an earlier priority date can be recognised for the portion of the claim relating to the specific widget is a critical consideration in determining if Z can be poisonous.

Partial priority

Partial priority refers to the concept that a claim can have more than one priority date.  That is, some of the subject matter of the claim is entitled to one priority date but other subject matter of the same claim is entitled to a different priority date.  This results where some subject matter within the claim is disclosed in an earlier priority application, and so is entitled to the priority date of that application, but other subject matter is not disclosed in the earlier priority application, and so is not entitled to that priority date.  Partial priority therefore allows for the same subject matter in related applications to have the same priority date, thus averting potential self-collision of related applications such as Y and Z above.

However, partial priority can be a complicated issue, especially in circumstances where the alternative embodiments having different priority dates cannot easily or succinctly be expressed separately and clearly.  Questions have revolved around whether, in order for partial priority to be acknowledged, embodiments entitled to different priority dates need to be explicitly separated[6] or whether disclosure of those separate embodiments can be implicit.

What’s the situation at the EPO?

The poisonous priority debate in Europe began several years ago.[7]

European patent claims can have more than one priority date.  Indeed, according to EPC Art 88(2), “multiple priorities may be claimed for any one claim”.  The right of priority covers only those elements which are disclosed in the application(s) whose priority is claimed.[8]

Decision G2/98[9] of the EPO’s Enlarged Board of Appeal determined that a claim could have more than one priority date, “provided that it gives rise to the claiming of a limited number of clearly defined alternative subject-matters.”  This appeared to introduce a restriction on the circumstances in which multiple priority dates for a single claim would be recognised.  That is, G2/98 suggested that, in order for multiple priority dates to be recognised for a single claim, there is a strict requirement to formulate the claim so as to explicitly recite alternative forms of the invention having different priority dates.  Unfortunately, there are circumstances under which this can be difficult and the requirement for “a limited number of clearly defined alternative subject-matters” is at the heart of the poisonous priority problem.

Diverging lines of case law developed, one following the strict requirements of G2/98 and the other following an alternative, more lenient, approach.  The more lenient approach to multiple priorities for a single claim, which is arguably a more pragmatic approach, acknowledges that a single claim can have multiple priority dates but recognises that it is possible to mentally split the subject matter without needing to explicitly recite the alternatives in the claim.

The divergent case law led to a referral to the EPO’s Enlarged Board of Appeal (case G1/15).  The first question referred to the Enlarged Board was:

  1. Where a claim of a European patent application or patent encompasses alternative subject-matters by virtue of one or more generic expressions or otherwise (generic “OR”-claim), may entitlement to partial priority be refused under the EPC for that claim in respect of alternative subject-matter disclosed (in an enabling manner) for the first time, directly, or at least implicitly, and unambiguously, in the priority document?

In its decision, G1/15,[10] the Board answered, “No,” and as further clarification ordered, “No other substantive conditions or limitations apply in this respect.”

Many European patent attorneys are breathing a sigh of relief because G1/15 essentially eliminates the strict requirements established by G2/98 and there is hope that this decision will put an end to the issue of poisonous priority for European patents and patent applications.

What about Australia?

Patent claims in Australia can also have more than one priority date.  Different forms of the invention within the same claim are treated as separate claims for each form of the invention and so are entitled to different priority dates.[11]

The approach of the Courts in Australia to the question of when there is a “different form of the invention” has not yet become clear.  In AstraZeneca v Apotex[12] a strict approach was adopted, the Court finding that the Patents Act 1990 did not intend to treat every potential variant of the defined invention as a different form of the invention.[13]  In contrast, in Nichia v Arrow Electronics[14] the Court concluded that two selection steps in the relevant claims resulted in different forms of the invention which were entitled to different priority dates.[15]

Thus it is difficult to predict for a particular case whether or not a strict approach to partial priority or a more lenient Applicant-friendly (or Patentee-friendly) approach will be taken in Australia.  It will be interesting to see if future decisions are influenced by the outcome of G1/15.

And New Zealand?

For applications filed in New Zealand on or after 13 September 2014 and PCT applications which entered the national phase in New Zealand on or after that date, the relevant legislation is the Patents Act 2013.  Patent applications filed or entering the national phase in New Zealand before 13 September 2014, and divisionals of those applications, are examined under the Patents Act 1953.

The Patents Act 1953 included a “prior claiming” provision[16] under which, in the context of later-published prior art entitled to an earlier priority date, only the subject matter of the claims of that prior art could anticipate.  In contrast, the Patents Act 2013 provides for “novelty only” prior art in its definition of the prior art base.[17]  According to the Patents Act 2013, all of the information (not just the content of the claims) contained in an earlier-priority-later-published application is relevant for novelty.  This change in the law represents an effort to harmonise New Zealand patent legislation with patent law in other jurisdictions (such as Europe and Australia).

While in theory a poisonous priority issue could arise under the 1953 Act, in practice this is more likely to arise under the 2013 Act because the entire contents of the earlier disclosure can anticipate.

The Patents Act 2013 is relatively new legislation so from a case law point of view it’s still early days.  Nonetheless, under the Patents Act 2013, a New Zealand patent claim can have only one priority date.[18]  If the subject matter of the claim is supported by the matter disclosed in an earlier application, the priority date of the claim is the date of filing of that earlier application (or, in the case of more than one earlier application, the earliest of them).  Otherwise, the priority date of the claim is the date of filing of the later application.[19]  This leaves the door wide open for poisonous-priority type attacks when divisional applications are involved and Applicants should stay alert to the potential dangers.

It’s worth mentioning that the “anti-self-collision” clause, which was present in the draft legislation, was removed before the Patents Act 2013 was passed.[20]  With no proposed legislative amendment currently on the cards, it seems that the issue of poisonous priority is unlikely to go away any time soon.


In Europe it is hoped that decision G1/15 has buried the poisonous priority problem for good.  In Australia, case law is still developing, whereas in New Zealand the issue of poisonous priority is alive and kicking.