Continuations-in-Part (CIPs) and Priority Claims

By Dennis Crouch

This post considers the meaning of priority in a continuation-in-part application and, in particular, when the claimed priority document can serve as prior art against the child.

Yesterday, I wrote about the majority opinion in Santarus v. Par Pharma (Fed. Cir. 2012). In that case, the patentee was fighting for the validity of its continuation-in-part (CIP) patent. A CIP is a patent filing that claims priority to a previously filed parent patent document but that also adds some amount of new material to the disclosure and/or claims. In CIP applications, priority date is determined on a claim-by-claim basis. Of course the priority date is important because it largely defines the scope of prior art available to potentially invalidate the patent claims.

In its decision, the Court of Appeals for the Federal Circuit ruled that some of the disputed claims were sufficiently disclosed in the parent filing and therefore properly claimed priority to that original filing date. Another set of claims, however, included matter that was added in the CIP. Those claims with new matter were not given priority to the original filing and thus had to rely on the filing date of the CIP application. Thus, for the claims with new matter any patent issued or document published more than one-year before the CIP filing date would count as prior art under 35 U.S.C. 102(b). See Paperless Accounting v. Bay Area Rapid Transit System, 804 F.2d 659, 665 (Fed. Cir. 1986).

At this point in the analysis, there is a major point of contention between the judges on one particular alleged prior art reference – whether the original patent filing counts as prior art against the child. The parent patent (really, the great grandparent patent) was filed in 1996 and issued as a patent in 1998. The descendent patent in being litigated in this case was not filed until 2003 but claims priority back to the original 1996 filing through a series of continuations and CIPs. Under normal circumstances, a 1998 patent would clearly serve as prior art against a 2003 filing under Section 102(b) as discussed above. However the system of priority claiming to parent applications is designed to overcome that problem by treating the child applications as if they had been filed on the same day as the original ancestor – so long as the procedures of 35 U.S.C. § 120 are followed. Interesting for the purposes of this discussion, Section 120 discusses priority at the patent level rather than claim-by-claim and asks whether the “application for patent” satisfies the written description and enablement requirements.

35 U.S.C. 120 Benefit of earlier filing date in the United States.

An application for patent for an invention disclosed in the manner provided by the first paragraph of section 112 of this title in an application previously filed in the United States . . . shall have the same effect, as to such invention, as though filed on the date of the prior application. . .

In Santarus, the invention of only some of the claims were sufficiently disclosed in the original specification. For those claims that lack priority, the big question is whether and to what extent the original patent can be used as 102(b)/103(a) prior art as part of an obviousness rejection.

In a per curiam opinion, the majority (Chief Judge Rader and Judge Moore) held that – since priority is determined on a claim-by-claim basis – that the original patent was prior art against the claims that lacked priority. Writing in dissent, Judge Newman argued that the priority should be given with regard to material disclosed in the priority document. Judge Newman’s approach also breaks the patent apart but divides priority more precisely according to which material was found in the original disclosure. Under Judge Newman’s approach, the original filing cannot be used as a priority document.

Newman writes:

The panel majority forgets that “matter disclosed in the parent application is entitled to the benefit of the filing date of the parent application.” Waldemar Link, GmbH & Co. v. Osteonics Corp., 32 F.3d 556 (Fed. Cir. 1994); see Litton Sys., Inc. v. Whirlpool Corp., 728 F.2d 1423 (Fed. Cir. 1984) (“The earlier filing date of the parent application pertains to material in the C-I-P application also disclosed in the prior application. 35 U.S.C. §120.”). Instead, the panel majority relies upon the common subject matter from the ‘737 [original] patent disclosure to invalidate the ‘885 claims supported by that subject matter. This is incorrect, for the common subject matter in the ‘885 patent is entitled to the ‘737 filing date. That entitlement is not lost by issuance of the ‘737 patent.

Of course, majority did not “forget” as Judge Newman suggested. Rather, the majority found that the patentee had “waived any argument that the ‘737 patent is not prior art to the other claims” by failing to appeal that point.

+ + + + +

Although distinguishable, the most on-point decision relating to this issue may well be In re Chu, 66 F.3d 292 (Fed. Cir. 1995). In that case (and with Judge Newman on the panel), the Federal Circuit ruled that the asserted priority document could be used as prior art because all of the claims included added limitations not found in the priority document. The court wrote:

Chu is entitled to the benefit of the Doyle patent filing date only if the Doyle patent discloses the subject matter now claimed by Chu. This, however, is admitted by Chu not to be the case. In fact, Chu states that “the invention as now claimed[ ] was not described in the [Doyle] patent.” Specifically, Chu concedes that “nothing in Doyle suggests that SCR catalyst be placed inside the bag filter.” Therefore, independent claim 1, which includes this limitation, and dependent claims 2, 4, and 14, are not supported by the Doyle patent disclosure. Accordingly, Chu cannot obtain the benefit of the Doyle patent filing date for these claims and the Doyle patent was properly relied on as prior art.

The obvious difference between the present case and Chu is that some of the claims are supported by the priority document.

93 thoughts on “Continuations-in-Part (CIPs) and Priority Claims

  1. 93

    Ned what you just wrote will be music to the ears of all those senior practitioners in the UK who were familiar with the provision of the UK Patents Act of 1949 (Section 6 I think it was) that took out of the state of the art exactly the “subject matter” of your own prio filing, should it occur after your prio filing ie during the Paris year. As you say, you ought to be able to publish to the world what you have in your prio filing, and still get valid claims to obvious variants when you “complete” your filing one year later. But mainland Europe had no experience of Section 6 so declined to write it into the EPC. Section 6 is one of a long shopping list of cherished national provisions (some from Germany) that didn’t make it into the super-slim, super-fit EPC.

    No great harm done, as far as I can see after 30+ years of the EPC.

  2. 92

    Max, all your have to do is read your post to understand that declaring your claim to lime unpatentable over a reverence that discloses what you have disclosed in your priority application makes no sense.

    If you file your provisional application, go to a conference, disclose its subject matter, an a third party publishes a review, just why should all that published material make your claim to limes obvious.  What have you withdrawn from the public domain?  Nothing.  Nada.  Zilch.  

    I complain because the basic idea behind prior art is not present.  it is there to prevent you from claiming that prior art, or obvious variations.  But when you yourself have priority to an earlier application, and the source of all that intervening art may just well be your own disclosure, then denying you a patent on a variation of something you invented makes no sense whatsoever.  None.

  3. 91

    Ned, do I have it as you would wish:

    First, your US prio claims citrus and has an enabling disclosure that includes a lemon embodiment and an orange embodiment.

    Your PCT filing, a year later, features not only lemon and orange but also an additional embodiment, lime, that delivers surprisingly better results.

    Thus, that generic claim to citrus in your EPO filing is in the prio too, and supported in the prio by lemon and orange examples. So that citrus claim in your PCT ought to get the US prio date. Disclosures after the prio date are ineffective to dispute its validity.

    Not so the claim specific to the lime embodiment. That one does NOT get the prio date. So intervening publications (by you or a 3rd) are, given Europe’s zero grace period, good for validity attacks on it. But, as we earlier agred, if the performance of limes really is unexpected, your limes claim is not only new but inventive.

    Otherwise, if limes perform no better than lemons or oranges, i should think your limes claim ought to be held obvious over that intervening disclosure.

    None of this is remotely difficult. Why are you making such a meal of it?

  4. 90

    The source…?

    I think you have a constitutional issue then, since the constitution plainly tells you which branch of the government is the source of patent law.

  5. 89

    Max, if my convention application claims the "citrus fruit" and relies on the disclosure of limes to support the genus.  Does the intervening disclosure of the examples from my priority case by a third-party anticipate that generic claim in the EPO convention filing?

  6. 88

    Ned, you write: “What I am talking about is where I disclose oranges and lemons in my priority application, the intervening prior art is oranges or lemons, and I later discover limes and disclose limes in my convention application filed in the European patent office. Can I add limes to the list of examples in my dependent claims in the EPO?”

    and once again I do not understand your difficulty.

    You tell me that a disclosure of limes is in your EPO filing but NOT in your US priority filing a year earlier. Of course you can claim what you disclose in your EPO filing. But any such claim (to limes) is NOT entitled to the priority date now is it. That’s as plain as a pikestaff, isn’t it?

    You tell me that surprisingly good limes are patentable over mere oranges or lemons. OK then. Your limes claim is good over the oranges and lemons art made public during the Paris year.

    In Europe, it’s all soooooo simple. What’s so difficult for you? I don’t get it.

  7. 87

    Max, I think finding that limes are much better than any other fruit for the application could get a patent over a prior disclosed genus even in the United States.

    Also, the lime example is not an example of something I am talking about.  What I am talking about is where I disclose oranges and lemons in my priority application, the intervening prior art is oranges or lemons, and I later discover limes and disclose limes in my convention application filed in the European patent office.  Can I add limes to the list of examples in my dependent claims in the EPO?

  8. 86

    In-house, I know.

    The real problem here is double patenting, not withdrawing from the public domain subject matter the public has a right to.  That is the reason for denying a patent based on prior art.

  9. 84

    Yes Ned. Of course. Standard situation. Here goes.

    In your prio you disclose and claim your invention: Give sailors citrus fruit to ward off scurvy. Your worked examples are oranges and lemons. You claim them too, in dependent claims. A year later you follow up, with PCT.

    Unbeknown to you, Dr Stephen Maturin is on the case too. In your prio year he also files for the same invention but adds that limes are surprisingly much better. In a dependent claim he claims using 100% limes.

    The British Admiralty requires that every warship shall carry a ton of limes. Limes are big. In your PCT you include a dependent claim to limes.

    The EPO examines your case and Maturin’s. Your claim 1 gets through. You were first to file for citrus. You’re rich.

    But Maturin is rich too. He was first to file for limes. His limes disclosure is earlier into the EPO than your limes claim. It deprives your limes claim of novelty. Sorry.

    But never mind. The British Navy is infringing your valid citrus claim, and Maturin’s limes claim too.

    Of course, none of this happens under the AIA, which is FtF on speed, turbo-charged FtF because the mere event of your filing your prio renders Maturin’s later-filed limes claim obvious.


  10. 83

    Max, have you ever seen an independent claim allowed and a dependent claim rejected over a reference whose date is later than the filing date of the priority document?

  11. 82

    Ned – I would want to file to remove the litigator’s argument that the genus claim is 1) not supported for the un-named species I am now trying to assert the patent against or 2) not supported across all members of my genus.

    Plus, if I have a patent to the specific species, I wouldn’t have to assert the broader genus patent. I may have better chances of surviving obviousness challenges in litigation.

  12. 81

    anon, exactly…That is why so many of the Federal Circuit decisions…

    Ned, you’ve taken my comment about the Supreme Court and turned it into a rant against the Federal Circuit.

    Please don’t do this. In my view, it is the Supreme Court, much more so than the Federal Circuit, that mucks up patent law; 180 degrees from your statement.

  13. 80

    Lost me there Ned. Sorry.

    At the EPO, dependent claims are examined, but only to the extent necessary. For example, suppose the independent claim is directed to “An ABC thing characterized by a D”. Now suppose dependent claim 2 recites……wherein the C is a K. Never fear. Claim 2 is going to get examined (and rejected).

    In post-issue opposition proceedings however, attention does focus on the independent claim. If it’s good, well so too are the claims dependent on it. If it’s bad though, there’s no need to look at any dependent claim (because the mere presence of the bad independent claim renders the whole patent bad). Better then have an “auxiliary request” (fall-back position) lined up, ready for that eventuality.

    Glad you think Europe operates “the very same principle” but I don’t yet see what exactly that Principle is. Sorry.

  14. 79

    Leo, in further reply, I think you have it just right. If you are effectively narrowing the scope of the claimed invention by adding additional features, then there really is no policy problem. The problem would arise if one were claiming a broader version of the same invention, or a different invention entirely.

  15. 77

    In house, if one has generic claims, why would one even need to file on a species?


    I thought this interesting from a case discussion why we file CIPs:

      In addition, “there are no strict time limitations for determining whether continued refiling of patent applications is a legitimate utilization of statutory provisions or an abuse of those provisions.” Symbol, 422 F.3d at 1385. The Federal Circuit has explained that there are legitimate grounds for refiling a patent application, such as to file a divisional application in response to a requirement for restriction, to present evidence of unexpected advantages of an invention when that evidence may not have existed at the time of an original rejection, or to add subject matter in order to attempt to support broader claims as the development of the invention progresses. Id.”

    SYNQOR, INC v. ARTESYN TECHNOLOGIES, INC., Dist. Court, ED Texas 2011

    Thus, adding species in order to support an eventual genus is important. Under the principles of Stempel and Symbol, this is entirely legitimate.

    References that show only what has already been filed should be “removed” by the priority claim.

  16. 76

    anon, exactly.

    It helps if one actually applies some intelligence to patent law. That is why so many of the Federal Circuit decisions are so bad. They write broadly on the narrow case, when in other contexts, what they have written doesn’t make much sense.

  17. 75

    anon, assume further anyone withing the skill in the art could find U given the disclosure of A, E, I and O:

    Then, I think, the first filer/inventor can claim G, but cannot claim U (no written description).

    The inventor of U gets nothing unless he can show unexpected results over G.

  18. 74

    IANAE, obviousness irrelevant to priority?

    A discloses and claims X + Y.
    B discloses and claims X + Z.

    How does one determine whether they are claiming the same invention?

    Ever hear of phantom counts?

  19. 73

    How about validity or pirate owners?

    (just kidding)

    Actually, may obviousness actually have a place in the discussion?

    If the intervening party cannot obtain a patent on a difference because that difference is obvious, may the obvious difference fall under the scope of the first patent?

    From KSR:

    Granting patent protection [to a second inventor, in our hypo, U]… may, in the case of patents combining previously known elements, deprive prior inventions [in our hypo, the first inventor, G+] of their value or utility.

    What was the value or utility of the prior inventions that was being threatened in the minds of the Supreme Court Justices based on the concept of obviousness?

    Granted, this is a stretch, but it is a colorable one.

  20. 72

    IANAE – Thanks for the discussion.

    I just think we run into a problem where the parent patent with the generic claims (as you correctly stated) “cover(s) products that are developed by competitors after the filing/invention date.” But, if the inventor actually tried to draft a claim to cover those later products, such a claim would “neither [be] described nor enabled, so it may not be claimed.”

    Anyone else see this as a problem?

  21. 71

    IANAE, can we make one more assumption please:


    Obviousness is irrelevant to whether the claim is entitled to priority, or to the related question of new matter. Obviousness has no place in this discussion.

  22. 70

    Genus A + B ??

    Can we reset the hypo thusly;

    Genus: G
    Disclosed and enabled species: A, E, I, and O
    Claimed in parent app: A, A + E, A + I, A + E + I, etc.
    Intervening application by another, disclosed and enabled species U


  23. 69

    In house, I think the law should, at a minimum, remove all references disclosing only A + B when you claim A + B + C. Double patenting is the problem, if anything.

    If someone else comes up with C, then the issue is dicier. Like a interference, the real question is who invented A + B + C first, who had possession.

  24. 68

    IANAE, can we make one more assumption please:

    C is obvious over the genus A + B? Obvious at the time of filing of t1.

    I assume a third party cannot get a patent here. Right?

  25. 67

    If my claim is supported and does recite the genus of all “vowels” (known or otherwise) “U” should be within the scope of the parent patent, right?

    It’s within the scope of your claim, yes. That’s not a problem. All patents are (or should be) intended to cover products that are developed by competitors after the filing/invention date. All patents would be worthless if they didn’t.

    How does one reconcile the policy

    One was described and enabled in the priority document, so it may be claimed. The other was neither described nor enabled, so it may not be claimed.

    If you want it in policy terms, I guess it’d be something along the lines of you have to disclose an invention at a particular date to be entitled to that date.

    Hypothetical: You file at time t1 describing and claiming “genus” A+B, with no mention of C. Someone else files at later time t2 for “species” A+B+C. You decide you’d rather be the one with the patent to A+B+C, so you file at later time t3 and claim priority to t1 because A+B+C is a kind of A+B. Who gets the patent to A+B+C?

  26. 66

    IANE – No, I don’t see the obvious problem with that. That’s why I asked.

    The reason (as I see it) that the inventor couldn’t have claimed the species “U” in the parent is because it wasn’t expressly stated in the parent. Some examiner’s would issue a 112. However, the examiner granted my claim to the genus. If my claim is supported and does recite the genus of all “vowels” (known or otherwise) “U” should be within the scope of the parent patent, right?

    How does one reconcile the policy that a claimed genus may be supported in an application by providing a representative number of examples against the policy that allows an examiner to rejected a claim for lack of priority when one tries to claim a non-listed species of the genus?

  27. 65

    Inventor couldn’t claim this in the parent because it would have failed WD.

    I personally think the Examiner should give the CIP priority to the parent

    You don’t see the obvious problem with that?

    “Inventor couldn’t claim this in the parent” is precisely the reason to not give the claim priority to the parent.

    Whether or not there remains a double-patenting issue is irrelevant.

  28. 64


    I actually favor Ned here. My experience with CIPs is when the patent discloses and claims a genus and you now want to claim a species that was not expressly identified in the parent. For example:

    1-Method of treating a disease with a vowel.
    (only A, E, I,and O disclosed. Vowel “U” not know at the time of filing the parent application.) Application publishes.

    “U” now identified as a vowel, and is publicly known. So, inventor files a CIP and claims:

    1-Method of treating a disease with vowel “U.” Inventor couldn’t claim this in the parent because it would have failed WD.

    I personally think the Examiner should give the CIP priority to the parent and issue a DP rejection, which I am happy to overcome with a terminal disclaimer.

    What does the peanut gallery think?

  29. 63

    If the reference shows a <b>new</b> combination that I am claiming in the CIP, I have to show full support for that new combination in the parent.  But if the reference shows only the old material, or only adds to the old material conventional items, it is removed, or at least, it should be.
    I think the law on this point is as clear as mud. 
    Regarding forever claiming obvious modifications … well that is double patenting.  The PTO does a very good job on that, IMHO. 


  30. 62

    At the risk of continuing the excessive use of a word that is already overused on this blog, I think you might be conflating the separate processes of determining the effective date for a claim and determining the universe of prior art, when you mention 112 support for the subject matter that is being removed.

    The effective date of a claim in a CIP has nothing to do with the subject matter in a 3rd-party reference. It’s a two-step process: determine the effective date of each claim in the CIP, and then determine whether references (including the parent application) qualify as prior art under sections 102 and 103. We don’t selectively remove just parts of intervening references.

    It sounds like you’re saying that a patent owner should be able to claim obvious variants of his originally disclosed subject matter forever (well, 20 years). When the later claims are narrowed versions of what was originally claimed, then that’s probably harmless. But what about obvious variants of the originally claimed material that aren’t covered by the original claims? That material, which is not explicitly described in the parent case, doesn’t become available to the public for 20 years?

  31. 61

    And, BTW, Leo, the issue is important not so much for filing new applications, or ever for prosecution, but for reexaminations and litigation where the prior art asserted is the same as that supported in a parent, as in the present case.

  32. 60

    Agreed that priority is determined on a claim-by-claim basis, generally speaking.

    Analyis at the level of elements (as opposed to “claims as a whole”, whatever that is supposed to mean) is certainly also “permitted”, however. As a practical matter, if an element of claim is not in a parent application, it does not matter what is in the rest of the claim. Priority for that claim cannot be obtained back to the parent.

  33. 58

    Yes, 112, p.1 support for "the subject matter" that is both being claimed and removed.  
    The problem involves "red herrings."  If I claim an improved carburetor and the reference discloses a carburetor, it should matter not that I claim the improved carburetor for a car, when a car was a well known application. 
    Claim 1. Carburetor.
    Claim 2. Carburetor for a car.
    Claim 3. Claim 1, for a car.
    Reference: Carburetor.


  34. 57

    The simple question, is whether one can receive an EFD for subject matter disclosed in a parent in a continuation for a claim to that subject matter such that an intervening disclosure of that subject matter by third parties is removed.

    The simple answer is yes, Ned, as you say. But isn’t the harder part determing whether the claimed material in the CIP was actually disclosed in the parent? Do we not require 112 support?

  35. 56

    Leo, in cases not being a statutory bar, one can say that the material cited in another publication is the applicant’s. Stempel also allow one to show prior invention of that subject matter disclosed in a third party reference.

    The simple question, is whether one can receive an EFD for subject matter disclosed in a parent in a continuation for a claim to that subject matter such that an intervening disclosure of that subject matter by third parties is removed?

    I think yes. So does Newman.

    IMHO, the Feds have conflated double patenting considerations with “prior art.”

    I am going to take a closer look at the new statute to see what it says as well.

  36. 55

    Leo [Filing a CIP] is almost never necessary. It might serve to confuse an examiner; you can decide for yourself whether that’s ultimately useful or not. There are a few situations where a CIP can be helpful, but they are rare. At least in my field. Those situations might be more common in other fields.


  37. 54

    Ned if everybody knows that a car has 4 wheels then “car with 4 wheels” is implicit in an enabling disclosure of “car”.

    Now, with that, is “Car with 4 wheels” novel over “car”? Yes or No?

    Or how about “car with 5 wheels”. Is that implicit in “car”? Or might it be an obvious variant of “car”? Yes or No?

    Now, what were we talking about? Patentability of A+B given an earlier claim to A. Not easy, I think you’ll agree. So, we need clear Rules of Play, right?

  38. 53

    Ned, no, that’s not how it works in Europe.

    In Europe (as you know) only stuff available to the public is available to build obviousness attacks. You know, I happen to think that it makes no sense to call something objectively obvious on the basis of stuff that no PHOSITA has yet had a chance to know.

    In Europe we use novelty to resolve priority between rival filers. Suppose each enabled in their docs as filed that which they respectively disclosed to the PTO, and claimed. Suppose none of these rival filers revealed anything to the public till dates later than the last of their respective filing dates at the PTO. Thus, on that last filing date, the public still knew nothing of the subject matter claimed by any of them. Shall they all get their claims through to issue? Well, no. The enabled disclosure content of each filing can be used to attack the novelty of the claimed subject matter of all of the filings that were made on dates later than that of the said enabled disclosure content. So, you get to issue everything except that which lacks novelty reelative to all those who filed earlier than you did.

    For me, nothing else makes sense.

    And this EPC scheme has the merit that it forces everybody to be very clear about i) the difference between novelty and obviousness and ii) what exactly is the subject matter of the precise individual claim under scrutiny.

    By the way, have no truck with the nutty concept of “extended novelty” half way between novelty and obviousness, espoused by German patent litigator (and Germany Govt adviser) Heinz Bardehle. It’s his wheeze to make more work for his litigation firm.

  39. 51

    Max, keeping this simple:

    A discloses and claims X + Y.

    B, later, discloses and claims X + Z.

    In the US, if Y and Z are obvious variations of X, A gets the “phantom” claim to X, to X + Y, and B gets nothing.

    Once you understand that X is the real invention, one can see just why the addition of obvious variations should not be rejected over one’s own priority document disclosing X.

    Ditto any interevening references discloses X.

    Anything else does not make sense.

    Is that how it works in Europe?

  40. 50

    Organic corn, I'm going to first make the assumption that A + C is a patentable invention with A as prior art.  Clearly, this combination is the "invention" of the third-party who first disclosed it.  I would think it would be proper to reject any later application claiming the combination of A + C.  Therefore, that third-party reference disclosing and claiming should be available against any claim that claims either A + C, or any other combination.

    However a proper "defense" should be that C was known at the time of the parent application and that the combination of A+ C is was well obvious to those of ordinary skill in the art such that the A+ C combination and simply is a claim to A in a known context.  Thus the claim should be considered supported in the parent. 

    Circling back, I would suggest that a reference to be good prior art to the combination of A+ C must be in a patent or patent application, and that patent application or patent should actually claim it.  Otherwise the failure to claim the combination should be taken as an admission that the subject matter is old.

  41. 49

    Why CIP,

    You premise “CIP App: Filed more than a [y]ear after Parent publishes

    Why is the year material? (I am assuming that the CIP has a proper link of priority to the parent).

  42. 48

    You’re rejected over Parent Pub plus third party refs teaching B.

    Leads me to the question “why ever file an application as a CIP?”

    It’s almost never necessary. It might serve to confuse an examiner; you can decide for yourself whether that’s ultimately useful or not. There are a few situations where a CIP can be helpful, but they are rare. At least in my field. Those situations might be more common in other fields.

    It seems that the only benefit is that the priority claim removes the parent as prior art. It does not remove intervening prior art

    I suspect that this is a common belief, but this is one answer that can’t possibly be right. Go look at the statute – neither 35 USC 102(a) nor 102(b)differentiates between publications of others and publications of yours. (At least not yet.) The only way to remove either as prior art under these sections is to get an early effective filing date for your claim. But the question of effective filing date is completely distinct from the question of what qualifies as prior art. In other words, if the CIP claim gets the parent’s filing date, then BOTH the parent case and intervening references are disqualified as prior art. If the CIP claim gets the CIP’s filing date, then both are potentially prior art, if they meet the requirements of 102(a) or 102(b). (But not 102(e), usually, due to 103(c).)

    If possible, please distinguish between what IS current law and what you think it SHOULD be under your theory or Newman’s

    I’m trying to describe the actual law. I’m not terribly interested in what it should be; practicing according to what the law should be doesn’t pay nearly as well.

  43. 47


    Good discussion points here. Leads me to the question “why ever file an application as a CIP?”

    It seems that the only benefit is that the priority claim removes the parent as prior art. It does not remove intervening prior art

    My hypothetical

    Parent App: Discloses and claims A

    CIP App: Filed more than a hear after Parent publishes, add B, which is arguably an obvious addition to A

    For my hypothetical, there are no other prior art references disclosing A, though there are references that arguably make adding B to something like A obvious

    Do I get a claim to A+B in the CIP, or am I rejected over Parent Pub plus third party refs teaching B?

    Thanks for any responses. If possible, please distinguish between what IS current law and what you think it SHOULD be under your theory or Newman’s

  44. 46

    Ned (and anybody else who is interested, and Jim if he is still there):

    With any First to File regime within Paris/TRIPS/WTO you have to be fair as between three inventors who, unbeknown to each other and underived from each other, and yet with interlaced filing and priority dates, all petition for issue of a claim directed to overlapping subject matter X, the Applicants comprising:

    Filer A, who claims the benefit of an earlier filing in a Paris Con country

    Filer B, who relies upon a cascade of national filings, and

    Filer C, who also filed at about the same time and, during prosecution, having seen A and B’s 18 month publications of their respective applications as filed, writes into his app, as a prosecution amendment, a claim to X.

    The prize can go to only one of them, namely the one who was first with a written description of X and an enabling disclosure how to put X into effect, the one who was first to plant the flag and then petition the PTO for exactly that piece of “land”, and before the prior art universe rendered that “land” old or obvious. As those 18 month publications issue, they add to the prior art universe. Between the rival filers at the EPO, each has to be novel over the other two, to succeed.

    Thus it is, that the notion of what is the disclosure content of any given document has to be always THE SAME, whether it is for the purpose of i) the doc is the earlier filing that is the basis of the priority date or ii) the doc is being used to attack novelty of a claim or iii) the doc is the app as filed and one must decide whether a prosecution amendment to that doc adds subject matter to it.

    G3/93 and its progeny say no more or less than that. Ned I think your Germans want it both ways, cake and eat it too. They want a liberal assesssment of whether they get benefit of their prio date while at the same time a strict view whether the docs of any other filer prejudices their novelty. Likewise, they want a liberal claim construction when asserting it and a strict one when defending its validity. Ned, this is muddying the waters deliberately. It is what litigators do. It ain’t right, is it?

    Shall we keep things simple and fair, or not?

  45. 45

    I agree with your 1 and 2, assuming “any prior art” in 2 means “any prior art other than the parent pub(s) of the CIP”

    However, your proposed point 3 is debatable. For example, a parent teaching A, a child teaching A+B, a intervening reference teaching A+C, can the intervening reference be used as 103 with another reference teaching B against the child? After all, a 3rd party reference almost always teaches something else other than “only subject matter which is supported”.

    For example, dupont claims teflon the chemical, a truely original invent. 3 months later 3M claims teflon aerosol, 3 months after that, dupont tries claims teflon gel.

    PTO have following choices:
    (a) use parent app of teflon and combine with any gel reference.
    (b) use 3M app of teflon spray, and combine with any gel reference.
    (c) allow teflon gel, no matter how obvious gelatinization is.

    Which one is okay under your point 3?

  46. 44

    Michael, the practice can be simplified as follows:

    1. A parent pub is never prior art to CIP properly claiming priority to it. (A patent granted on such might be relevant for double patenting grounds.)

    2. The examiner could cite any prior art disclosing any element of the claim if the prior art date was before the CIP filing date.

    3. Just as now, the applicant would have the burden of showing support in the parent for what he is claiming, but element by element. If the reference discloses only subject matter which is supported, it is removed in the same manner it is now under Stempel.

  47. 43

    Consider the hypothetical:

    Application 1 discloses and claims A, and is published (Parent Pub).

    More than 1 year after the Parent Pub, Application 2 is filed (as a C-I-P to application 1) and discloses A+B, where B is not supported by Application 1. Application 2 has a single claim to A+B.

    A third-party publication (3rd Party Pub) discloses A and publishes after the filing date of Application 1 but more than 1 year before the filing date of Application 2.

    There appear to be three approaches to this situation:
    I. The one taken by the majority: the Parent Pub and the 3rd Party Pub can both be used against Application 2.
    II. The Parent Pub can not be used against Application 2, but the 3rd Party Pub can be used against Application 2.
    III. Neither the Parent Pub nor the 3rd Party Pub can be used to teach feature A in Application 2.

    I think Judge Newman (and Ned Heller) are arguing for approach III, but it seems very difficult to apply in practice. First you have to parse the claim of Application 2 into supported and unsupported limitations. Then you have to subtract out the supported limitations from the teachings of the 3rd Party Pub before considering obviousness in view of the 3rd Party Pub.

    Approach II appears at first glance to be more fair, but leads to some illogical outcomes. For example, what if the 3rd Party Pub incorporates all of the Parent Pub. Why should the 3rd Party Pub be treated differently than the Parent Pub?

  48. 41

    So?  IN the hypo, the reference discloses A.  There is nothing in the prior art that discloses B.

    Now, assume A is an improvement in an old combination, let's say a carburetor in a car.  I claim a car with an improved carburetor.

    The reference discloses the improved carburetor.

    I file a CIP.  I claim a car with an improved carburetor. That claim remains patentable over the reference.

    I add obvious variations in dependent claims: Wherein the car has four wheels.

    Why is such a claim, in point of fact, not fully supported in the parent given that everyone knows that a car has four wheels?

  49. 40

    There may be a problem with your instant scenario.

    You have A filing a CIP after he has abandoned the application, thus breaking any chain of priority to A’s FTF first application (A’s second filing is not a true CIP at all).

    Perhaps you can reset your hypo?

  50. 39

    IANAE, if the question is who invented first, we look to the common subject matter between two rival applicants.  Who ever gets that priority also gets priority with respect to obvious variations.  That is the principle involved here and what Newman is really arguing.

    in FTF, Inventor A discloses and files on A, but abandons the application.  B later discloses and files on A.  He also lists a number of obvious variants.  I hope you would agreed that B not only cannot patent A, but he should not be able to patent the obvious variants.

    A then files a CIP, claims A, and the obvious variants that were first disclosed by B.

    While we should all agree that the CIP should have priority WRT A, the closer question involves claims to the obvious variations first disclosed by B.  My first thought is that there may be a problem here.

  51. 38

    The EPO case was G3/93 – an opinion of the Enlarged Board of Appeal – not a decision.

    Opinions have greater force than decisions, despite normally having no one who cares presenting any useful arguments. It’s the way we do things in Europe.

  52. 37

    Why is the priority claim no longer good?

    Because the claim “A with variation B” is not a claim to A. More to the point, it is not a claim to anything that you’ve disclosed in any priority document.

  53. 36

    IANAE, of course, but am I contending anything different?

    Parent A.
    CIP claims A. The reference discloses A. The benefit is good.

    CIP adds to A, B. It claims A. It claim A with variation B. The reference discloses A. Why is the priority claim no longer good?

  54. 35

    I don’t see the problem.

    The problem is that he’s getting a priority date for a claim that is not entitled to the priority.

    Patent term isn’t always the point. In some technologies you don’t need or even want 20 years.

  55. 34

    I contend that my priority claim is good against all references showing A. What is your view?

    My view is that your priority claim is good against all references for a claim reciting A. See the difference?

  56. 33

    Max, glad to, but I need your help, just a bit, as I have not looked into this for a bit. I recall a German article that was discussing convention priority and the change in the rule regarding this that occurred in the mid ’90s. It was an EPO case that essentially held that a priority document was good prior art against any claim that did not have 100% support in that document.

    Once I have the number of that case, I can conduct a Google search on the German discussion of the case.

  57. 32

    IANAE, remember, we are not talking about references that disclose subject matter different from than that which is disclosed in the parent. The reference discloses only that which is disclosed in the parent. Who invented that subject matter? That is the question. Clearly the patentee invented that subject matter. So long as the inventor is not extending patent term, and so long as he is not extending the scope of his claimto include subject matter invented by others, then I don't see the problem.

  58. 31

    If I disclose and claim A, and a reference discloses A, I can file a 131 affidavit swearing behind the reference. I can further show that I have the benefit of a priority claim to an earlier application that also discloses A.
    Now the question becomes whether I can add dependent claims to an independent claim claiming A, having established my priority to the A subject matter. I contend that my priority claim is good against all references showing A. What is your view?

  59. 30

    Ned you keep on telling us that the Germans are in open revolt. If it’s “open” where can I read about it?

    I guess you are alluding to the very recent shift in German Supreme Court thinking on novelty, bending the knee to EPO jurisprudence after a 30 year hold out. Indeed it does come hard, to a German lawyer, to acknowledge that pan-European law is anything other than German law, writ larger.

    Share it with us please.

  60. 29

    you are arguing double patenting.

    It may well be double patenting also. It is possible for the same claim to have two defects.

  61. 28

    The question is, “Who invented A?”

    Who first filed to A. And even then, that’s only the question if the claim is to A. If the claim is to B, even if B is an obvious variation of A, then the question is “who invented B”, and a previous filing disclosing A is no answer at all.

  62. 27

    Yes, provided the new subject matter is invented before the publication date of the original (otherwise it is 102(a)-type prior art), you could avoid the parent qualifying as prior art under 102(b). However, for claims that rely on the new material, that would still be the case if it were to be filed as a new application and you kept the parent to deal with the old claims. If there’s double patenting, a terminal disclaimer could handle that. No need for a CIP; it doesn’t get you anything extra.

  63. 26

    Yeah, Mark, and for all that modernity, we no longer need the rule you propose for excluding priority for common subject matter. Patent term is not being extended by adding obvious variations via CIPs. Moreover, even under the prior regime, patent term was not extended due to double patenting rejections and terminal disclaimers.

    While what you say may be the current thinking of the Federal Circuit and the EPO, it was not always the law. It is modern the product of misguided thinking by shallow thinkers.

    The Germans are in revolt right now. They think, and rightly so, that if a priority document discloses a bicycle, and a claim in convention filing is to a green bicycle, an obvious variation, that claim should not be invalid over the priority document.

    There is no good reason for the Federal Circuits position. It is illogical and it flies in the face of common sense just as argued by Judge Newman.

  64. 25

    Fish, we all know that they do provide priority with respect to the new matter. The question is much more complicated when one considers the effective priority of common subject matter.

    What do you think about this:

    Parent discloses A.
    CIP discloses B.
    Refernce before CIP but after Parent discloses A.

    I claim:

    1. A
    2. The invention of claim 1, plus B.

    Analysis please.

    Is A patentable over the reference? Clearly yes.

    But what about variation claimed in claim 2?

  65. 23

    bad, I think you miss the point.

    The reference discloses A, the same subject matter that the priority application discloses. That is the topic of conversation.

  66. 22

    An examiner, I urge you to read In re Stempel to understand what we are talking about.

    The parent discloses A. The reference discloses A, the very same thing. Not B, Not C. A

    The question Newman is arguing is the priority date for a CIP also disclosing A good for A to eliminate the reference that also discloses A?

    The question is, “Who invented A?”

    The patenting of obvious variations in continuations is a double-patenting problem, not a prior art problem.

  67. 21

    No good reason… How about this: Inventor X discloses and patents invention ABC during copendency files a continuation X-1 to keep application alive.

    1 year to the publication of the patent a publication Y not by inventor X disclosing ABC and D is disclosed.

    Inventor X than files a CIP off of CON X-1 disclosing ABC and D after publication Y and over 1 year from the publication of the original patent.

    In your position Inventor X could get a patent on ABCD even though publication Y was submitted before the CIP.

    How can the public really be sure that Inventor X really invented ABCD? It could have been derived from publication Y or even if it was independly invented publication Y may have been first.

    Why should anyone perform work on invention ABC for the next 20 years when their work can be then patented by inventor X?

  68. 19

    excluding the priority document from prior art consideration for all claims.

    I don’t think it’s sound policy to allow an applicant to essentially arbitrarily designate any of his prior filings as “not prior art” when they’re published and a statutory bar to the world. Not even at the expense of patent term.

    If you invent something that is not entitled to an earlier priority date, you have to deal with the prior art as you find it at that time. If it was obvious over your previous filing, you should have thrown it into the earlier spec at the time.

  69. 18

    Every analysis of patentability of claims (novelty, obviousness, sufficient disclosure, etc) is for the invention “as a whole”. Likewise, as for priority (and consequently eligible prior art), one cannot obtain priority for part of a claim, dissecting a claim into what is old and what is new. The asserted priority claim must be evaluated for the claim as a whole.

    The invention set forth in a particular claim, as a whole, either is fully supported by a parent (grandparent, etc) application’s description or it isn’t. If a particular claim relies on subject matter added to the disclosure in a CIP, then that particular claim is not entitled to the benefit of the parent application’s filing date and it is as if that claim were presented in a new application. Everything prior to the actual filing date is prior art, including possibly the inventor’s earlier work. The inventors of an improvement don’t get special treatment, just because they also invented the earlier more basic technology.

    CIPs are basically a relic from the days when applications weren’t published until the patent was granted, the patent term was calculated from the patent grant date, and there were 102(e)/103 benefits to having the old and new claimed matter in a single application, particularly when the inventorship entity for the old and new material overlapped but was not identical. Now that applications are routinely published at 18 months (with a few exceptions) and the patent term is calculated from the oldest domestic application priority claim, CIPs have limited continued usefulness.

  70. 17

    I usually agree with Newman, particularly on policy, but I think that she’s wrong here.

    When the parent published more than 1 year ago, ANYONE should be able to come up with and patent an improvement. Her policy would allow ONLY the patentee to patent (much less bring to market) improvements during the 20 years of the first patent. It’s essentially a renewable monopoly with a cap of about 40 years as long as you keep an application pending. That’s entirely too long for promoting innovation and competition.

  71. 16

    It is just amazing how many experienced patent attorneys do not understand how CIP’s work and how they differ from regular continuations

  72. 15

    Regarding: Invention = Claim

    I think that is right in terms of how the Patent Act is ordinarily interpreted. So, I think you have a pretty good argument that the statute permits claim-by-claim analysis of priority.

  73. 14

    IANAE, I think that you have a good point here. Although I think that the system could also make sense if the law took the additional step of excluding the priority document from prior art consideration for all claims.

  74. 13

    And, let me just add this, for as long as provisional applications and priority applications existed in Europe, for almost 200 years, the rule was that the priority document eliminated intervening references disclosing the same subject matter as described in the priority document. The priority document could also not be used against the referring document. This regime made sense. In re Stempel was to the same principle and in accord.

    It was been the Federal Circuit which changed this basic regime into a regime where one looks at whether a claim is “fully supported” in the priority document. That’s 100% flipped from the law as it existed in the world for centuries.

    The cases it relied on did not support that proposition. One of the cases had this fact pattern: old lamp plus vanadium was disclosed in the priority document, which was also published more than a year prior to the filing date of the CIP. The referencing document disclose additional materials besides vanadium. The CCPA held that the additional materials did not find support in the priority document, therefore the intervening publication could be used against claims that depended upon the additional matter. This of course makes court a lot of sense, because the claim was the use of a particular material, vanadium, in an admittedly old application. It was the vanadium that was the point of novelty, not the old application.

    The principles of In re Stempel are also embodied in the literal words of the Paris convention.

  75. 11

    Dennis, you should know that the Germans are having a significant difficulty with this doctrine when claiming priority to earlier applications. Following US law, the Europeans held that a claim not fully supported in the priority document did not have priority to that document. Thus the priority document itself could be used as prior art against a later patent application.

    The Germans are in open revolt. They fundamentally do not understand why if one has a claim to subject matter A that is fully supported in the priority document, why someone cannot add other subject matter to that claim or claim variations.

    Thus I invent a bicycle and disclose a bicycle in my priority document. I claim a bicycle in my convention application. If I add that the bicycle is green and green is not supported in the parent, under law, I lose priority with respect to the green bicycle, and the green bicycle becomes unpatentable. To the Germans this makes no sense. To me it makes no sense. To the common man on the street it makes no sense. And apparently to Newman it makes no sense.

    If you add into the mix policy considerations about extending the term of a basic invention by adding obvious subject matter in continuations, then one sees that this really is a policy issue that is more pertinent to double patenting and not to prior art. So long as one can not extend the term of the basic invention beyond 20 years from its filing, there is no good reason not to recognize the priority date of the basic invention even in CIP’s.

  76. 9


    See my post on the Higher Written Description thread at 9:16 AM. I think you are over-reading the case in making an element-by-element priority stand. Newman herself talks of claims that deserve priority.

  77. 8

    Anon, I think Newman’s thinking is that the subject matter supported in the parent is entitled to the filing date of the parent. That is not an unreasonable interpretation of section 120.

  78. 7

    IANAE, one can overcome such an intevening reference today if it is not a statutory bar under Stempel. I am sure you knew that.

    The only think Newman is suggesting is that we extend Stempel to the to the case of statutory bars as well.

    The primary policy reason we did not do that in the past is that adding obvious variations to a basic invention through CIP’s could potentially extend the life of the basic invention of exclusivity forever if one was measuring a term from the date of issue. But that policy reason vanishes when we measure term from the date of the earliest priority claim in the United States.

  79. 6

    I fundamentally do not see anything wrong with this even in the statutory bar context,

    I have a problem with dissecting a claim and assigning different priority dates to different parts of it. Whether or not one part happens to be in a parent claim that is entitled to an earlier date.

    If you have a claim to A and B, and the A part is entitled to priority, what of intervening art that makes it obvious to add B to A, but without explicitly teaching A?

    Priority doesn’t make any sense unless it’s all-or-nothing for each claim.

  80. 5

    Professor Crouch,

    Don’t you think the phrase “as to such invention,” in 35 USC 120 carries with it the meaning of “per claim?” (per 35 USC 112 para 2)

  81. 4

    Lew, I think the issue is a little bit more subtle than you suggest. The question is whether In re Stempel should be extended when a reference as more and a year old.

    Stempel will allow the priority benefit to a date of invention of subject matter A against an a reference showing A in a claim that adds B.

    I fundamentally do not see anything wrong with this even in the statutory bar context, although I would require a terminal disclaimer via an obviousness type double patenting rejection.

  82. 3

    Dennis, I would also add that the majority opinion appeared to invalidate the claims even though the majority found that they were fully supported in the parent. I don’t recall a single claim where the majority did not find that they were so supported. This is what I found odd, and what Newman also found odd.

    The problem appears to be based solely on the alleged failure by the appellant to raise the issue that the parent was not properly prior art to claims supported in the parent. Once they do not deserve this issue on appeal, they were deemed to have waived it.

    I think that is what set Newman off more than anything else.

  83. 2

    The majority decision is consistent with Tronzo v Biomet, 156 F3d 1154 (Fed Cir 1998), which held that publication of the foreign counterpart of a US parent application is available as 102(b) prior art against the claims in a CIP that weren’t entitled to benefit of the parent. Judge Newman, in a concurring opinion, disagreed for the same reason as her dissent in Santarus.

  84. 1

    Dennis, close, but not quite there.

    Parent: A, publication is more than a year prior to CIP

    CIP: A + B;

    Newman: A has priority of Parent. A is removed as a reference against A + B. It is B that does not have priority to the Parent.

    The problem with Newman’s system is that allowing such a claim in a term-from issue date system based on the novelty of A is that the claim to A might be extended indefinitely by adding B, C, D, E….. But this could be remedied by double patenting and it normally is.

    In a term-from filing regime, the concern about extending term vanishes completely.

    Regardless if there was a good reason in the law for not according the A subject matter its priority data before 1995, that reason seems to be gone. In truth, the patentee invented A, and should reasonably have priority to it.

    Consider Stempel. Parent/reduction invention date discloses A. CIP: Claim is to A + B. Reference is A. 131 affidavit can overcome the A reference.

    The policy reasons behind Stempel are sound.

Comments are closed.