The Relevance of Invention Date in Patent Prosecution: Part III

This is Part III of my series on the relevance of the invention date in patent prosecution. You may download my entire paper here. This part considers the role of provisional patent applications as an incomplete but functional substitute for invention-date rights. 

* * * *

With some caveats, I argue that provisional patent applications can serve as functional substitutes for reliance on the U.S. filing grace period. Provisional patent applications have become popular since being made available in 1995. The USPTO reports that applicants filed 143,030 provisional patent applications in fiscal year (“FY”) 2008, compared with 466,147 non-provisional utility applications that year. U.S. applicants file a great majority of these provisional applications Consequently, the provisional filings correspond to more than half of the annual total of utility applications filed by U.S. residents.

A provisional application is essentially a low-cost placeholder that allows an applicant to establish a priority date by filing a description of the invention along with paying an administrative fee to the USPTO. The applicant may later file a non-provisional application claiming the benefit of the provisional application’s filing date. The provisional patent application has a twelve-month non-extendable pendency. During that time-period, the application is kept secret and is not examined. Thus, to rely upon the provisional priority date, the applicant must submit a non-provisional application within one year of the provisional filing date. Professor Seymore notes several additional strategic advantages of filing provisional patent applications: “extending the patent term to twenty-one years, postponing the examination period, trolling for prior art, and avoiding an allegation of inequitable conduct.”

To explore the relevance of provisional patent applications as a tool for claiming priority, I created a dataset of 5000+ provisional patent applications filed between January 1, 2001 and October 1, 2007. Of those, over half (52%) were still secrete as not yet identified in the priority claim of any publicly available non-provisional application. Although some of these will eventually become public as associated with a non-provisional filing, in the end, close to 40% or more of the provisional application filings will never be linked to an issued patent or public non-provisional application.   Collectively, these figures indicate that a large number of provisional applications are themselves left floundering to be abandoned twelve months after filing. The large number of abandoned provisional applications suggests to me that provisional patent applications are likely being used as early place-holders to cheaply secure a priority date, rather than as a means of extending the patent term or postponing examination. In a separate study looking at the 620,000+ issued patents for which applications were filed between 2000 and 2005, I found that seventeen (17) percent of the patents in the data-set claim priority to at least one provisional application. On average, the patents claiming priority listed 1.4 provisional patent applications. In a separate study, I found that on average abandoned non-provisional applications tended to claim priority to 2.0 provisional patent applications. This combining of multiple provisional applications to support a single patent supports this theory that provisional applications as a substitute for filing date evidence. Notably, each provisional can serve as evidence of an incremental advance in place of a corroborated laboratory notebook. Anecdotal reports suggest that this practice has become especially prevalent among universities developing early stage innovations. However, the value of incremental provisional application filings is limited by the hard twelve-month deadline for filing a non-provisional application.

For an inventor who has conceived a new invention, the legal benefits of filing a provisional application, rather than relying on the invention date, are compelling. If the filing deadlines are met, a provisional patent application will serve as conclusive evidence of priority—without any need for any affidavits or declarations—so long as the provisional application discloses the invention claimed in the non-provisional application “in the manner provided by the first paragraph of section 112.” In other words, the specification of the provisional application must “contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable” an ordinarily skilled artisan to practice the invention claimed in the non-provisional application.

Unlike an applicant seeking to assert a prior invention date, an applicant claiming priority to a provisional application need not provide corroborated evidence—or indeed any evidence—of actual reduction to practice, conception date, or reasonable diligence in working toward reduction to practice. Rather, the USPTO and courts treat the filing of a provisional patent application as a constructive reduction to practice, thus allowing the applicant to rely on the simple evidence of the paper filing.

An oft-stated disadvantage of provisional application filings is that they may create a false sense of security among patent applicants. Certainly, a hastily-filed provisional patent application may fail to properly describe the invention to be claimed in a later-filed non-provisional application. However, the validity of this criticism does not suggest that applicants should merely rely upon the hope that they will possess sufficient evidence to antedate references based on a prior invention date. Rather, risk-averse applicants should quickly file a well supported application either in provisional or nonprovisional form.

17 thoughts on “The Relevance of Invention Date in Patent Prosecution: Part III

  1. 17


    We have discussed this on several threads, but this is the first one I stumbled upon, so I’ll post my comments here.

    Hal Wegner puts out a daily email type of “blog”. Today’s piece features a commentary on the newly proposed “24 month provisional”. There is an interesting piece that relates to our discussion on full disclosure, which I provide in a non-substantial quip:

    Failure of the 24 month de facto provisional application scheme:

    The commercialization efforts involving disclosure of the invention to
    outsiders within the first year from filing will create a fatal defect for claims
    presented at 24 months from the filing date which lack a full “written description”
    in the provisional application. The New Railhead case explodes the myth that
    claims can be presented based upon a provisional that does not meet the
    requirements of 35 USC § 112, ¶ 1.
    New Railhead Mfg. v. Vermeer Mfg., 298
    F.3d 1290 (Fed. Cir. 2002)(Michel, J.).

    Thus, “[c]laims enjoy the earlier filing date only if the provisional application
    provided adequate written description under 35 U.S.C. § 112, ¶ 1.” Trading Techs.
    Int’l, Inc. v. eSpeed, Inc., 595 F.3d 1340 (Fed. Cir. 2010)(Rader, J.)(quoting New
    Railhead, 298 F.3d at1294). In other words, claim limitations that would be added
    for the first time 24 months after filing without support in the provisional
    application as filed would not be entitled to priority. Where a claim is presented
    after the priority filing, the applicant must show “that he or she ‘had invented each
    feature that is included as a claim limitation[.]’” Cordis Corp. v. Boston Sci. Corp.,
    561 F.3d 1319, 1332 (Fed. Cir. 2009)(Dyk, J.)(quoting New Railhead, 298 F.3d at

    This is the point that I have been trying to beat into you. You do need full disclosure to the claimed invention. Quick and Dirty provisionals will fail you.

    Thanks Hal.

    (Oh, and btw, NAL posts one of her usual four page journal articles over at Gene’s site: link to if you want to be put to sleep)

  2. 16

    Tara, IMO yes — if the prior art is not a statutory bar.

    But, we have the issue of 102(c) and (g) abandonment to consider.

  3. 15

    Who would know the answer to the question, would a well written provisional, where a non-provisional was filed much more than one year later, support the proof of the invention date of the provisional under the first-to-invent-doctrine in the US? While the USPTO would not accord priority as having not complied with the one year requirement, does the application still needs to note that invention priority date? Thanks

  4. 14


    You again in this thread neglect to address the one key word: claims

    I need not dispute Pfaff v. Wells – that case does not neglect claims, as you consistently do.

    The train is rolling in Ned, you gotta pick up the baggage if you want to hop aboard.

  5. 13

    ping, then you dispute Pfaff v. Wells?

    I have drawings enabling one to make and use. I file them as a provisional. I then “sell” the invention per Pfaff and later make a delivery. Perhaps I also describe the invention in some “white paper.” Later, but within a year of the provisional, I file my application.

    Now, if I could demonstrate a prior reduction to practice, you would assuredly agree that I could swear behind my own “sale,” public use and public disclosure. Why, then, should I not be able to swear behind my own sale, etc. to show the drawings even if they are not a formal reduction to practice if it is the existence of the drawings that determines that the invention is “complete.”

    Assuming I could file the drawings in a Section 131 affidavit as proof of prior invention, and I submit that one can do that after Pfaff v. Wells, what difference does it make if the drawings are on file in the USPTO?

  6. 12

    Simple bald statements that one must have full section 112, p.1, support for a claim to prove prior invention cannot be right.

    So says the bald man…

  7. 11

    proof of prior invention is all that is needed.


    How is that prior invention defined? What is the prior invention? The only thing defining the invention are the claims – sorry, but all the baggage required to support the claimed invention has to be carried on the train if you are going to make that trip.

  8. 10

    EG, New Railhead, IIRC, did involve a statutory bar. Full Section 112, p.1 support was required for this reason.

  9. 9


    You’re correct, I did overstate the 35 USC 112, first paragraph issue as I was looking at the ability to rely on the provisional as the priority date for the subsequent non-provisonal. The Federal Circuit was pretty explicit it New Railhead Manufacturing v. Vermeer Manufacturing that you better comply with 35 USC 112, first paragraph in the provisional if you were to rely upon it for a priority day in the non-provisional:

    “However, for the non-provisional utility application to be afforded the priority date of the provisional application, the two applications must share at least one common inventor and the written description of the provisional must adequately support the claims of the non-provisional application.”

    Whether the provisional might otherwise be relied upon as evidence of the conception of the claimed invention (as of a certain date) is an entirely differnt matter.

  10. 8

    I disagree that provisionals need to fully comply with Section 112, p.1 for all purposes. Clearly they do to establish an effective filing date. But, effective filing dates are primarily useful in establishing the critical dates for statutory bars. When a reference is not a statutory bar, proof of prior invention is all that is needed. Proof of prior invention, in turn reqires only proof of possession, and under the principles of Pfaff v. Wells, this should require only drawings sufficient to enable one to make and use the invention.

    It is quite apparent we need a full discussion on this point because there appears to be a substantial misunderstanding of the legal issues involved. Simple bald statements that one must have full section 112, p.1, support for a claim to prove prior invention cannot be right.

  11. 7

    Dennis –

    Characterizing a provisional as a “cheap substitute” reflects an incorrect understanding of the practical realities, both legal and business.

    To have any value (or rather, any non-destructive value!), a provisonal has to meet all the requirements of § 112 ¶ 1. To get there almost always requires involvement of a patent attorney (in the 10+ years that provisionals have been permitted, I’ve had one inventor who came to me with a document that would have been adequate as a provisional.) Most inventors come to me with marketing brochure level disclosures, which are usually adequate for “how to use,” but totally inadequate for “how to make” or best mode. Even that one document had all the “must” and “must not” statments that are in bad specifications; these had to be removed before I could file the provisional.

    Plus — the filing fee is totally irrelevant, even the attorney fee is only moderately relevant. The biggest cost is the inventor’s time to prepare an adequate disclosure. The inventor is usually the key person in the company. Taking about a day of his/her time to get him to prepare some drawings, give me an interview, answer some follow up questions, and review the document before filing (I make mistakes! Oh how I need the inventor’s review!) is often a really crucial bite out of the company’s resources!

    And then — to force this cost on the company for an invention that turns out to be worthless because the new Patent Reform Act and “forced to file” requires fast filing, without a year to think and weed out the turkey inventions — that adds insult to injury.

  12. 6

    “A provisional application is essentially a low-cost placeholder.”


    You characterize provisionals exactly the same way I do. Unfortunately, too many don’t know or simply ignore the fact that provisionals must also comply with the first paragraph of 35 USC 112. That means the provisional is only as good a “placeholder” for that which you describe and enable (and the provisional will be a completely empty “placeholder” if you fail to describe the “best mode”). Nothing wrong with proviisionals, which can be quite useful, but you’ve got to realize their limitations.

  13. 4

    “the myth of a cheap provisional patent application as a substitute for a well-written patent application”


  14. 3

    “[P]rovisional patent applications can serve as functional substitutes for reliance on the U.S. filing grace period.”

    That seems like an odd way to look at it, since the primary purpose of a provisional patent application is to establish the earliest possible effective filing date for whatever you describe and enable in the provisional (well that, and effectively getting an additional year of patent term). So I don’t see how a provisional can be viewed as a “functional substitute” for reliance on the grace period. In some ways, Professor, you seem to be blurring the distinction between bar dates and the concept of priority of invention.

    As for the disadvantage of creating a false sense of security, that is primarily the fault of patent attorneys who continue to peddle to inventors the myth of a cheap provisional patent application as a substitute for a well-written patent application (be it provisional or non-provisional). In most instances, about the only cost savings for a provisional should be the reduced filing fee for a provisional, and, perhaps, using informal drawings prepared by the inventor (although even there I have seen lots of provisionals with drawings so poor that they affect the adequacy of disclosure).

  15. 1

    Raising then the question of what happens when the non-pro contains a more/ better/ greater scope (of) invention or inventions than does its pro; in other words, the pro doesn’t fully support the non.

    Was there any value then to filing the non?

    Has the non become a de facto CIP?

    How would this play out in litigation?

Comments are closed.