Looking Beyond Blazemarks on Trees — It’s Time to Revisit the Description Requirement in the Wake of Warner-Jenkinson

1997Journal of the Patent & Trademark Office Society, volume 79 (1997), pages 625-42


by Harris A. Pitlick[1]

Now that Warner-Jenkinson[2] has highlighted the critical importance of documenting on the record in an application the reasons for a narrowing amendment, I believe it is an appropriate time to revisit the law on new matter/description requirement, since how much a claim is narrowed to avoid the prior art may very well determine whether that claim survives the application of prosecution history estoppel in a doctrine of equivalents infringement case.[3] Consider the following scenario.

Inventor-applicant White originally discloses and claims as his broadest claim a composition comprising A, B, and 10-30% of C. White also discloses a preferred range for C of 15-20%, and examples wherein the amount of C is between 15% and 20%. This is the only disclosure regarding the amount of C. Because of prior art rendering unpatentable a composition comprising A, B, and 28% or more of C (but which does not render unpatentable compositions containing a smaller amount of C), White amends the claim to recite A, B, and 10% to less than 28% of C, but the examiner rejects the claim as failing to comply with the description requirement of 35 USC 112. White, believing applicable precedent supports the examiner, or at least is sufficiently vague that reversal on appeal is problematic,[4] . . . ultimately amends the claim to recite A, B, and 10-20% of C, and receives a patent thereon. Black begins selling a composition comprising A, B, and 24% of C. White sues Black for patent infringement.

Under applicable precedent, Black may be home free, if the prosecution history estops White from enforcing his claim beyond the 20% maximum that White was required to insert in order to obtain his patent.

Arguments available to Black under present law are that as a competitor, he was reasonably entitled to conclude that White gave up subject matter inclusive of Black's composition in order to obtain a patent,[5] that limits imposed by prosecution history estoppel can be broader than those imposed by the prior art such that prosecution history estoppel can preclude assertion of a range of equivalents sufficient to encompass an accused product, even if the assertion of such a range is not precluded by the prior art,[6] and that by acquiescing in the rejection instead of appealing it, White forfeited his right to challenge its correctness now.[7] Black should also argue that White should not be able to obtain an exclusionary right through the doctrine of equivalents on subject matter he could not literally claim because he was not in possession of that subject matter as of his filing date.

Arguments available to White under present law are that as a competitor, Black was not reasonably entitled to conclude that White gave up subject matter inclusive of Black's composition in order to obtain a patent,[8] that a narrowing amendment to obtain a patent does not always mean that the patentee is estopped from recapturing any of what was originally claimed; that the amount of recapture, if any, depends on the circumstances,[9] and that the narrowing amendment was not made to avoid the prior art, since all White was required to do to avoid the prior art was to recite an amount for C only slightly below 28%.[10] White should also argue that prosecution history estoppel should not apply because it was only the description requirement of 35 USC 112 that prevented him from claiming the higher maximum amount for C.

Who prevails? As far as I have been able to determine, such a case has yet to be decided.[11]

Could White have succeeded on appeal if he had maintained his position that the limitation "below 28%" was in compliance with the description requirement? Should White have succeeded?

Before these questions are answered, some history and discussion.

The notion of the description requirement as separate and distinct from the enablement requirement of 35 USC 112 may be analogized to a UFO, first identified in 1967,[12] fifteen years after launch,[13] and confirmed by the early 1970s.[14]

It is well-settled that the test for compliance with the description requirement is whether, as an issue of fact, a person skilled in the art would reasonably conclude from the disclosure whose filing date is being relied on that the inventor had possession as of that date of the later claimed invention, and how the disclosure accomplishes this fact is unimportant.[15] I will refer to this test as the possession test. Moreover, it has become a mantra in almost every description requirement case that such cases are to be decided on a "case-by-case basis."[16] Satisfaction of the description requirement insures that subject matter presented in the form of a claim subsequent to the filing date of the application was sufficiently disclosed at the time of filing so that the prima facie date of invention can fairly be held to be the filing date of the application.[17]

That the description requirement is a separate requirement from the enablement requirement has been criticized as a misreading of the statute, most notably by then Chief Judge Markey of the CCPA.[18] My intent here is not to agree or disagree with this criticism; I assume the separateness of the description requirement is here to stay, at least until the Federal Circuit, in banc, holds otherwise. Rather, my focus here is to critique its past,[19] and hopefully improve on its future, application.

A good place to start is In re Wertheim.[20] In one of its many holdings, the CCPA held that the disclosure of a range of 25 to 60%, with examples of 36% and 50%, described "between 35 and 60%." The significance of cases like Wertheim, however, go beyond their particular facts. The case is important because of its explicit recognition that inventors occasionally discover that their contributions to the art are not as broad as they first thought. The court stated:

Inventions are constantly made which turn out not to be patentable, and applicants frequently discover during the course of prosecution that only a part of what they invented and originally claimed is patentable. As we said in a different context in In re Saunders:[21]

To rule otherwise would let form triumph over substance, substantially eliminating the right of an applicant to retreat to an otherwise patentable species merely because he erroneously thought he was first with the genus when he filed. Cf. (citation omitted). Since the patent law provides for the amendment during prosecution of claims, as well as the specification supporting the claims, 35 USC 132, it is clear that the reference to "particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention" in the second paragraph of 35 USC 112 does not prohibit the applicant from changing what he "regards as his invention" (i.e., the subject matter on which he seeks patent protection) during the pendency of his application. Cf. (citation omitted).[22]

The CCPA soon had occasion to apply the above in In re Johnson,[23] holding that a claim to a genus with a recital of a negative proviso that did not appear in the specification complied with the description requirement. The negative proviso had the effect of excluding from the scope of the claim two species originally disclosed in the specification as within the invention, and was inserted to avoid having the claims read on a lost interference count. The claim recited a formula O-E-O-E', wherein E and E' were both positively defined, and ended "with the provisos that E and E' may not both include a divalent sulfone group and may not both include a divalent carbonyl group linking two aromatic nuclei." The proviso literally excluded more than the two species.[24] The court stated:

The notion that one who fully discloses and teaches those skilled in the art how to make and use a genus and numerous species therewithin, has somehow failed to disclose, and teach those skilled in the art how to make and use, that genus minus two of those species, and has thus failed to satisfy the requirements of §112, first paragraph, appears to result in hypertechnical application of legalistic prose relating to that provision of the statute.[25]

A few months later, the CCPA had another opportunity to review "the sort of 'hypertechnical application' of the written description requirement of §112 which was recently criticized in [Johnson]" in In re Driscoll.[26] In holding that a disclosure of a group of chemical compounds containing a plurality of groups defined by Markush terminology, one of which groups contained 14 members, described the same group of compounds otherwise identical except that the 14-member group was now limited to only one of the members, the court stated:

Were the board's decision permitted to stand, future applicants, particularly in cases of this nature, would in all likelihood find themselves in the predicament reflected in the following observation by Judge Learned Hand:
If, when [applicants] yield any part of what they originally believed to be their due, they substitute a new "invention," only two courses will be open to them: they must at the outset either prophetically divine what the art contains, or they must lay down a barrage of claims, starting with the widest and proceeding by the successive incorporation of more and more detail, until all combinations have been exhausted which can by any possibility succeed. The first is an impossible task; the second is a custom already more honored in the breach than in the observance, and its extension would only increase that surfeit of verbiage which has for long been the curse of patent practice, and has done much to discredit it. It is impossible to imagine any public purpose which it would serve.[27]

It's hard to quarrel with the above passages. Indeed, it is doubtful that an inventor, ignorant of the precepts and nuances of patent law, after discovering that a species or a subgenus of a genus that he regards as his invention is old or obvious, would not believe that he possessed, at the time he invented the genus, that genus minus the old or obvious species or subgenus. In Johnson, the inventors were permitted to exclude the old or obvious species, ostensibly because they literally disclosed these species in their specification. The fact that Johnson actually excluded somewhat more than the two species was seemingly unimportant to the court, since this fact was relegated to a footnote that had no discussed bearing on the court's decision.

Clearly, the most ideal narrowing amendment to avoid patent-defeating prior art is that which accomplishes a "surgical" removal from the claim of unpatentable subject matter only. Anything more than a surgical removal puts the applicant at risk, like White in the scenario. The only obstacle which has stood in the way of such an amendment is the application of the description requirement, as the next case shows.

In Ex parte Grasselli,[28] the broadest claim on appeal was couched in Jepson form, reciting in a particular process using an oxidation catalyst "in the absence of sulfur and halogen, the improvement comprising using as the oxidation catalyst a catalyst wherein the atomic ratios of the elements are described by the formula Aa Bb Cc Sbd Ox, wherein A is phosphorus, boron, Ni, Co, alkali metal, alkaline earth metal or mixture thereof; B is iron, vanadium, manganese, chromium or mixture thereof; C is molybdenum, tungsten or mixture thereof . . . said catalyst being free of uranium and the combination of vanadium and phosphorus." Emphasis added. The board found that "the negative limitations" recited in the claim introduced "new concepts"[29] and held that the claim did not comply with the description requirement. The board amplified its holding, in what I will call "excess baggage", stating:

It might be added that the express exclusion of certain elements implies the permissible inclusion of all other elements not so expressly excluded. This clearly illustrates that such negative limitations do, in fact, introduce new concepts.[30]

Grasselli sought reconsideration before the board. In adhering to its decision, the board added to the excess baggage, stating:

Appellants . . . used the recitation "a catalyst wherein the atomic ratios of the elements are described by the formula." This recitation is, on its face, subject to more than one interpretation. . . . Appellants' claim recitation as to the catalyst, apart from the negative limitations, is not completely unambiguous on its face and could have been argued to inherently exclude other catalyst components. Appellants, however, chose not to so argue and in fact have expressly indicated their intention that the recitation should be open to all elements not expressly excluded by the negative limitations alluded to above. This position taken by appellants convinces us of the correctness of our original decision, wherein we held that the express exclusion of certain elements not discussed in the original specification together with the implicit (now virtually explicit) inclusion of all other elements not originally disclosed in the original specification constitutes new matter, i.e., lacks description . . . .[31]

What is immediately clear after reading both the original opinion and the opinion on reconsideration is that the facts on which the holding of Grasselli is based, to use the words of the board, "[are] not completely unambiguous on [their] face." What is clear is that Grasselli added at least some negative limitations to the claims in an attempt to overcome rejections over prior art. But the parts of the appealed claim emphasized above can be considered as two sets of negative limitations--one set as part of the Jepson preamble (no sulfur or halogen) and the other set as part of the improvement clause (no uranium or combination of vanadium and phosphorus). The board refers to "the negative limitations" (emphasis added). Were both sets added during prosecution or only the improvement clause limitation, or only parts of one or both? Only the prosecution history reveals the answer.[32]

The prosecution history reveals that the application in Grasselli was originally filed with no disclosed negative limitations, and no references to sulfur, halogen or uranium. Although the oxidation catalyst as originally disclosed was inclusive of embodiments wherein the combination of vanadium and phosphorus was present and embodiments wherein the combination was not present, none of the some seventeen exemplary oxidation catalysts disclosed contained the combination, nor was there any express disclosure, pro or con, regarding that combination. In each example, the inventive oxidation catalyst was combined with silica in a 80:20 ratio, respectively. The examiner finally rejected the claims, finding that both sets of negative limitations violated the description requirement. The examiner stated: "The concept of explicit non-presence of these agents cannot properly be said to be inherent in a lack of disclosure of the concept of explicit addition of these agents."

It is perfectly understandable why Grasselli asserted in their request for reconsideration before the board that the claim was open to elements not explicitly excluded. The board having raised the issue in its original opinion, Grasselli had to be concerned by an estoppel if they asserted otherwise (i.e., that the claim was not open to elements not explicitly excluded), since it might be difficult to later enforce the claim against the use of the same oxidation catalyst in combination with additional components, such as silica or equivalents thereof, or with additional non-disclosed elements, even if nonessential.

Nevertheless, since the above excess baggage was part of the facts relied on by the board in its holding, these facts must be considered when relying on Grasselli as precedent.

Grasselli argued in the PTO that the description requirement was satisfied, in that they merely restricted the scope of their claims to distinguish over the prior art. Grasselli relied on various precedent of the CCPA, including Johnson and Driscoll, before the PTO. The examiner distinguished the cited precedent.[33] The board adopted the examiner's position, stating: "The examiner's distinctions between the present case and the prior decisions cited by appellants are correct and we adopt his position in that regard as our own."[34]

Thus, the public record is unclear as to what distinctions over precedent such as Johnson and Driscoll the board relied on in Grasselli.

Unfortunately, Grasselli has been relied on by patent examiners for the broad proposition that claims containing negative limitations per se not appearing in the specification do not comply with the description requirement. Indeed, some examiners do not even distinguish between those excluded embodiments that had originally been disclosed, such as the phosphorus and vanadium therein, whose exclusion is permitted by Johnson, and those not originally disclosed, such as the uranium therein.[35]

Cases like Grasselli result because of the difficulty of determining precisely what the inventor possessed at the time in question. One commentator stated the following about the possession test:

[T]he "possession" standard strikes this writer as unduly vague and "subjective." The possession standard shifts the focus from those skilled in the art--the specification's audience--to the inventor. Enablement, the classic standard for the written description, may be thought of as a relation between the inventor and the public. The critical question is whether the inventor enables the artisan to practice the invention. The procedure and type of evidence that would permit this to be shown are familiar and relatively non-problematic. In contrast, the "possession" standard focuses on the inventor alone, and asks whether he "possessed" the invention. Thus instead of looking to what the disclosure enables those skilled in the art to do, it asks whether the disclosure permits them to detect something about the applicant--namely, his "possession" of the invention. But precisely how is this to be ascertained? The courts have answered: by asking whether the disclosure conveys to those skilled in the art that the inventor "possessed" the invention. But this only pushes the problem back one level; how is the skilled artisan supposed to recognize this?[36]

In situations where the claimed invention is broader than the invention literally disclosed, application of the possession test is a relatively easy, non-controversial exercise. But when the claimed invention is narrower, it is virtually impossible to identify precisely what an inventor possessed as of the date in question. Indeed, consistent with my comment about inventors above, I suspect most inventors of a genus would consider their invention to embrace all species and possible subgenuses therein. The best result to be hoped for in applying the possession test is to change the question--instead of asking what the inventor possessed, ask what invention should the inventor be credited for possessing.

Perhaps the results in description requirement, narrowing amendment, cases would be more palatable if the reviewing tribunals spent more time explaining the public policy reasons behind the description requirement. In Fields v. Conover,[37] one of the early description requirement cases alluded to above,[38] involving an interference, the issue was whether Conover's disclosure of a relatively broad genus described the subgenus claim or species claim that were copied from Fields' patent, where Conover disclosed no species within the terms of the copied claims. Conover's genus claims had already been allowed, and non-elected claims were pending which, if allowed, would dominate most, if not all, of the scope of Fields' claims. In holding that Conover did not comply with the description requirement, the CCPA stated, after acknowledging that Conover complied with the enablement requirement of 35 USC 112:

If Conover is allowed to copy Fields' claims merely because the application . . . points indistinctly and ambiguously in the general direction of that subject matter, the socially valuable incentive to further research and development provided by the opportunity to obtain subservient patents will be considerably diminished. (Emphasis added.)[39]

In effect, the court was focusing on how advancement of the art is potentially jeopardized by granting a patent to Conover, as opposed to the focus of the possession test, which is whether Conover possessed particular subject matter.

When the possession test is considered in light of the above-quoted passage from Fields v. Conover, a new test can be formulated as follows:

An inventor is entitled to a filing date for subject matter that he regards as his invention as long as granting that filing date does no harm to others' incentive to obtain subservient or other[40] patents.

I will call the above the incentive test. Let's apply this test to the following two scenarios, involving hypothetical inventor-applicants Green and Brown, respectively.

Green has originally disclosed and claimed a genus large enough so that each species included therein is not immediately apparent. During prosecution, prior art is cited disclosing species not literally disclosed in the specification and rendering only a relatively tiny percentage of the genus unpatentable. (The genus is otherwise patentable.) Due to the size of the genus it is impossible as a practical matter to amend the claim by positively reciting what is included, thereby implicitly excluding only those species rendered unpatentable by the prior art. Instead, Green seeks to amend the claim by adding a negative limitation that explicitly excludes the unpatentable part of the genus.

Brown has originally disclosed and claimed a composition for use X and consisting essentially of elements J, K, and L. Brown additionally disclosed that other elements may optionally be added, and provided some examples, but had no explicit disclosure about elements that cannot be added. Prior art is cited during prosecution rendering a composition consisting essentially of elements J, K, L, and M unpatentable for unrelated use Y. Brown did not disclose M as an optional element, although the presence of M in Brown's composition would not materially alter its basic and novel characteristics.[41] Let's also assume that M is relatively expensive compared to J, K, and L. Brown seeks to amend his claim by adding a negative limitation that explicitly excludes M.

How does the incentive test impact Green? If Green is permitted to insert the negative limitation he desires, does that diminish the incentive of others to find species or subgenuses within the scope of Green's now somewhat narrowed genus claim? I don't see how. The opposite side of the coin is that, unable to obtain a patent of scope he believes he deserves, Green now may have the incentive not to patent, thereby perhaps depriving the relevant art of his advance and frustrating the Constitutional purpose of promoting progress of science and the useful arts.[42]

How does the incentive test impact Brown? If Brown is permitted to insert the negative limitation he desires, does that diminish the incentive of others to find improvements--such as a composition consisting essentially of at least elements J, K, and L, and element P, which produces unexpected results over the composition not containing element P--of Brown's now somewhat narrowed genus claim? I don't see how here, either. While adding the negative limitation desired might invite potential infringers to add an amount of element M to a composition containing J, K, and L to avoid infringement, Brown is not concerned, since he considers the possibility unlikely, in view of the high cost of element M. If the only alternative amendment available to Brown is to change "consisting essentially of" to "consisting of", Brown, like Green, might also not have the incentive to patent, since such a claim would invite the addition of any optional element to easily avoid infringement.

Suppose Brown's twin, Browne, had disclosed element M as an optional element. Then, under the authority of Johnson, the negative limitation would pose no description requirement violation. Is there any reason to permit Browne, but not Brown, to insert the desired negative limitation, when it was only both fortuitous and serendipitous that Browne literally disclosed an optional element that happened to be disclosed by the prior art? I submit not.

Contrast the above Brown scenario with a scenario where, instead of deleting element M from the scope of the claims, Brown attempts to amend the claim by explicitly including element M. It is clear and noncontroversial that Brown would then fail both the possession test and the incentive test.

The incentive test is easier to apply than the possession test, since it does not require a determination of whether a narrower invention is part of the invention described, or a different invention.[43] It follows that the incentive test will yield a more predictable result. The questions one must ask in applying the possession test are: Did Green have possession of his original genus minus the extracted subject matter as of his filing date? Did Brown have possession of his original composition minus element M as of his filing date? Similarly, did Grasselli have possession, as of the filing date, of a process conducted in the absence of sulfur and halogen using as the oxidation catalyst a catalyst wherein the atomic ratios of the elements are described by the formula Aa Bb Cc Sbd Ox,, said catalyst being free of uranium and the combination of vanadium and phosphorus?

The answer to these questions may depend on through which end of the telescope the questions are viewed. Using Grasselli as an example, if one looks at the original disclosure only, it is virtually impossible to characterize their invention as the above-described process. On the other hand, if one looks at the claim first, and then the original disclosure, where every disclosed example is devoid of the excluded elements, it is a much simpler exercise to find that the invention included the above-described process. In Ruschig, the very first description requirement case, the CCPA indicated the former controls,[44] but this view of the possession test has not often (if at all) been repeated in later cases, and it does not appear to have been formally adopted as part of the test. Indeed, by focusing on the original disclosure only, it is the rare case where the description requirement will be found to be satisfied when literal supporting language does not exist.

The incentive test also provides a fairer result than the possession test. It rewards those applicants like Johnson, Green, and Brown (and Grasselli) who, consistent with the above-quoted passages from Wertheim, Johnson, and Driscoll, simply limit their unpatentably broad genus invention to a narrower genus that avoids the prior art, either by explicitly deleting subject matter or by explicitly narrowing the recital of the genus. At the same time, it denies a filing date to those who would overreach, like Ruschig and Conover, who have disclosed only a broad invention, and perhaps a few species, for subsequently invented, either by them or others, species and/or subgenus improvements.

Let's apply the incentive test to one last scenario.

Inventor-applicant Gray has disclosed and claimed a process for making Z using calcium. No substitutes for calcium are disclosed. Gray later presents a claim drawn to a process for making Z using magnesium. Calcium and magnesium are generally considered to be equivalent or interchangeable. Under present law, Gray's new claim would not comply with the description requirement since obvious variations of what was originally disclosed do not meet the possession test.[45] The incentive test would provide the same conclusion. Here's why.

The legal conclusion of obviousness requires that the invention be considered as a whole, and involves more than prima facie obviousness.[46] At the time Gray adds the new claim, there cannot possibly be sufficient facts to determine legal obviousness. At this point, according to what the prior art knows about magnesium's expected equivalence to calcium, magnesium is no more than prima facie obvious over calcium. Permitting Gray the new claim would harm the incentive of others to invent around the use of calcium. Indeed, the most likely invent around situations are those which involve expected equivalents, such as magnesium, with the hope of discovering some unexpected result. To the extent magnesium is legally obvious, Gray suffers no insurmountable harm from not being permitted to claim it, since Gray should be able to prevail under the doctrine of equivalents against an alleged infringer using it for making Z. To the extent it is not legally obvious, either because of unexpected results or other secondary consideration evidence, Gray cannot legitimately complain, because he did not appreciate the significance of the magnesium embodiment as of his filing date.[47]

In sum, the incentive test need not be characterized as a repudiation of the possession test. Rather, it can be considered as an adjunct thereto in that it helps assure that the inventor is credited with possessing an invention as of the filing date in question that is consistent with the Constitutional purpose of promoting progress of science and the useful arts.

So, where do we go from here? Hopefully, an applicant in the not too distant future with a fact situation like any of the above White, Green, or Brown scenarios will be able to successfully convince a reviewing tribunal--the board and/or (preferably) the Federal Circuit--of the soundness of the arguments raised herein enough to result in a precedential decision. What is needed are more decisions that discuss and demonstrate an understanding of the public purposes of the description requirement, with explanation of why the claimed invention does or does not further these purposes, and less resort to old metaphors.[48]

Meanwhile, some recent cases should be of help to patent applicants on description requirement issues. First of all, it is clear that any unjustifiably broad reliance on Grasselli that a negative limitation not literally disclosed violates the description requirement should have been answered by Ex parte Parks.[49] In Parks, the board held that the addition of the term "conducted in the absence of a catalyst", which did not appear in the original disclosure, nevertheless complied with the description requirement. The facts in Parks concerned a method involving decomposition of a sample to generate nitric oxide. The board distinguished Grasselli, stating "[u]nder the facts in that case, it was held that the negative limitation ["free of uranium and the combination of vanadium and phosphorus"] introduced new concepts in violation of the description requirement. . . ."[50] The board reviewed the disclosure concerning the decomposition step, none of which mentioned a catalyst. The board stated: "Throughout the discussion which would seem to cry out for a catalyst if one were used, no mention is made of a catalyst."[51] In holding that "it cannot be said that the originally-filed disclosure would not have conveyed to one having ordinary skill in the art that appellants had possession of the concept of conducting the decomposition step generating the nitric acid [sic, oxide] in the absence of a catalyst," the board cited two declarations in the record of an expert to the effect that one having ordinary skill in the art would have recognized that the reaction generating nitric oxide, according to the disclosure, is conducted without a catalyst.[52]

Note that in Parks, the board relied on declarations under 37 CFR 1.132. One might have believed after Parks that at least at the level of the board it was clear that such declarations could be probative of a description requirement issue during ex parte prosecution. Yet it seems from the opinion in the recent case of In re Alton[53] that it might not have been that clear. At any rate, Alton now leaves no doubt. The Federal Circuit held in Alton that it was error not to consider on its merits a declaration under 37 CFR 1.132 submitted on the factual issue of whether the description requirement was satisfied. Alton is thus a valuable precedent for inventors to provide evidentiary support under 37 CFR 1.132 before an examiner, especially in those cases where the specification alone is questionable on a description requirement issue.

Getting back to the first scenario, is there any public purpose that is frustrated by granting White a patent on a composition comprising A, B, and 10% to less than 28% of C? Can it be seriously urged that White did not have possession of a composition comprising A, B, and 10% to less than 28% of C as of his filing date? I submit not to both questions. At least until the dust of Warner-Jenkinson becomes more-settled, I would urge applicants and their attorneys to contest similar factual situations, since Warner-Jenkinson highlights the importance of doing so.

Of course, appeals are expensive. Avoiding such situations in the first instance makes the best sense. Warner-Jenkinson reminds us once again of the importance of disclosing a reasonably sufficient number of gradually narrower subgenuses in patent applications so as to minimize the potential prosecution history estoppel effect of a too narrowing amendment.[54]

First published in Journal of the Patent & Trademark Office Society, volume 79 (1997), pages 625-42.


[1.] Oblon, Spivak, McClelland, Maier & Neustadt. The views expressed here are personal and do not necessarily reflect the views of the firm or any of its clients. (Ed. Note: Between 1981 and 1994, the author briefed and argued the PTO's position in some one-hundred ex parte appeals before the Federal Circuit and its predecessor CCPA.)

[2.] Warner Jenkinson Co., Inc. v. Hilton Davis Chemical Co., 117 S.Ct. 1040, 41 USPQ2d 1865 (1997).

[3.] In Warner-Jenkinson, the Supreme Court held that in the absence of reasons recited in the file wrapper, there is a rebuttable presumption that a narrowing amendment during the prosecution of a patent application was made for "a substantial reason related to patentability. Id. at 1051, 41 USPQ2d at 1873. The Court remanded to the Federal Circuit for determination of the reasons for part of a narrowing amendment, i.e., for inserting the lower pH in the term "at a pH of from approximately 6.0 to 9.0", of a claim previously containing no pH limitation to overcome prior art suggesting a pH of greater than 9. On remand, in Hilton Davis Chemical Co. v. Warner-Jenkinson Co., Inc., ___ F.3d ___, 43 USPQ2d 1152 (Fed. Cir. 1997)(in banc), the Federal Circuit, in a per curiam opinion, directed the district courts in the future as follows: "If the district court determines that a reason not related to patentability prompted an amendment, the court must then decide if that reason is sufficient to overcome estoppel." 43 USPQ2d at 1154. Factors to be considered include "public notice and reliance on the prosecution history, as well as the need for fairness to the patentee." Id. The court remanded to the district court "to conduct an inquiry to ascertain whether Hilton Davis can rebut the presumption by showing the reason for the amendment of the claim . . . and whether that reason is sufficient to overcome the estoppel bar to the application of the doctrine of equivalents." Id. One commentator recently suggested, before entry of the Federal Circuit's decision on remand, that the term patentability, which was used by Justice Thomas in his majority opinion in some places along with reference to prior art in others, with regard to reasons for making amendments, is ambiguous, since at least the first two paragraphs of 35 USC 112 also address issues of patentability. Kenneth R. Adamo, The Waiting At The (Patent) Bar Is Over - The United States Supreme Court Decides Hilton Davis, 6 J. Fed. Cir. Bar 343, 359 (1996), available in 79 J. Pat. & Trademark Off. Soc'y 421 (1997). Adamo suggests that "[t]o obviate this issue, a better read of 'patentability' as used by the [Supreme] Court might be that afforded by the letter of Title 35, Chapter 10--Patentability of Inventions--includes only sections 100-105; Section 112 appears in Chapter 11--Application for Patent." Id. The Federal Circuit's decision on remand arguably muddies the water even more, since it suggests that prosecution history estoppel may attach even for amendments not related to patentability, without any discussion of the scope and meaning of patentability.

[4.] Compare, e.g., In re Blaser, 556 F.2d 534, 194 USPQ 122 (CCPA 1977) (3-2 decision) (literal disclosure of some amount up to 1.6 mols held not to describe range of 0.6 to 1.6 mols) with In re Voss, 557 F.2d 812, 194 USPQ 267 (CCPA 1977) (term "crystalline content . . . at least 50% by weight" not literally disclosed held to be described by literal disclosure of "glass-ceramic material" coupled with evidence that one skilled in the art would have attributed the recited crystalline content as inherent in that material). "Perhaps the most appropriate conclusion that a disinterested observer can draw from Blaser and Voss is that the determination of whether there is adequate description of the invention is frequently more subjective than the CCPA would care to admit." Edward C. Walterscheid, Insufficient Disclosure Rejections (Part III), 62 J. Pat. Off. Soc'y 261, 282 (1980).

[5.] See, e.g., Haynes International, Inc. v. Jessop Steel Co., 8 F.3d 1573, 28 USPQ2d 1652 (Fed. Cir. 1993), on rehearing, 15 F.3d 1076, 29 USPQ2d 1958 (Fed. Cir. 1994). In Haynes, three original claims to an alloy recited "20 to 24%", "about 21 to 23%", and "about 22%", respectively, of chromium. In response to a rejection over prior art disclosing a range for chromium of 14 to 26%, and a preferred range of 22 to 25%, patent applicant of the patent in suit submitted affidavits exemplifying only an alloy containing 21.96% of chromium, along with statements regarding criticality of the 20 to 24% range. The examiner finally rejected all the claims over the prior art and applicant appealed to the Board of Patent Appeals and Interferences ("board"). The board affirmed the rejection of the two broadest claims, but reversed on the narrowest claim. Applicant cancelled the two broadest claims and the patent thus issued with a claim reciting "about 22%" of chromium. Applicant never submitted additional data or otherwise pursued either of the two broadest claims. The accused product had a chromium content of between 20.74 and 20.81%. The court held that prosecution history estoppel applied and that the alleged infringer, as a reasonable competitor, could have concluded that the patentee, by cancelling the two broadest claims and failing to submit additional test data in a continuation application, gave up coverage that would have included the accused product. Id. at 1578, 28 USPQ2d at 1656.

[6.] Id. at 1579, 28 USPQ at 1656-57, citing Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 879, 228 USPQ 90, 96 (Fed. Cir. 1985).

[7.] See, e.g., Litton Systems, Inc. v. Whirlpool Corp., 728 F.2d 1423, 221 USPQ 97 (Fed. Cir. 1984). In Litton, the patent applicant Litton, after a new matter rejection on amended claims in a parent application, filed a CIP application to cover the subject matter of the amended claims, rather than argue the rejection, and the CIP issued as a patent. During litigation on the patent, Litton argued that the CIP application was unnecessary and that the additional subject matter contained therein was not new matter. The court held that Litton was estopped from making this argument. The court stated: "A patent attorney should not be able, however, to choose one course of action within the PTO with the anticipation that, if later checked, he or she can always choose an alternate course of prosecution in a trial before a federal judge." Id. at 1439, 221 USPQ at 107.

[8.] See, e.g., Zenith Laboratories Inc. v. Bristol-Myers Squibb Co., 19 F.3d 1418, 30 USPQ2d 1285 (Fed. Cir. 1994).

[9.] See Warner-Jenkinson Co., Inc. v. Hilton Davis Chemical Co., 117 S.Ct. 1040, 41 USPQ2d 1865 (1997). See also Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931, 939 n.2, 4 USPQ2d 1737, 1743 n.2 (Fed. Cir. 1987) and cases cited therein.

[10.] In the scenario, White made two amendments, the first one to overcome the prior art; the second to overcome the description requirement issue. If instead White had made only the second amendment, because he believed that the first amendment would have provoked a rejection over the description requirement, he would have been in a much more difficult position to argue that his amendment was not for reasons of avoiding the prior art.

[11.] The Federal Circuit has determined in a number of cases that amendments made for section 112 purposes only may or may not raise an estoppel depending on the specific facts of the case. See, e.g., cases discussed in Pall Corp. v. Micron Separations Inc., 66 F.3d 1211, 1220, 36 USPQ2d 1225, 1231 (Fed. Cir. 1995). However, no case I found presented facts similar enough to the scenario to be considered dispositive, nor did any appear to recognize that the various requirements of the first and second paragraphs of 35 USC 112 may each raise its own particular issues with regard to the application of prosecution history estoppel.

[12.] In re Ruschig, 379 F.2d 990, 154 USPQ 118 (CCPA 1967). In Ruschig, the CCPA was faced with an issue of whether the subject of Claim 13 therein (i.e., a species of organic compound--chlorpropamide--not literally disclosed), was described by a relatively large genus inclusive of the species and by other species. The court held that the description requirement was not satisfied. As the court explained:

It is an old custom in the woods to mark trails by making blaze marks on the trees. It is no hope in finding a trail . . . to be confronted simply by a large number of unmarked trees. Appellants are pointing to trees. We are looking for blaze marks which single out particular trees. We see none.

Id. at 994-95, 154 USPQ at 122. The court later stated:

Working backward from a knowledge of chlorpropamide, that is, by hindsight, it is all very clear what route one would travel through the forest of the specification to arrive at it. But looking at the problem, as we must, from the standpoint of one with no foreknowledge of the specific compound, it is our considered opinion that the board was correct in saying:
Not having been specifically named or mentioned in any manner, one is left to selection from the myriads of possibilities encompassed by the broad disclosure, with no guide indicating or directing that this particular selection should be made rather than any of the many others which could also be made.

Id. at 995, 154 USPQ at 123. Later, in response to Ruschig's argument that the enablement requirement of 35 USC 112 was satisfied, the court finally laid its cards on the table:

[T]he question is not whether [one skilled in the art] would be so enabled but whether the specification discloses the compound to him, specifically as something appellants actually invented. We think it does not. Second, we doubt that the rejection is truly based on section 112, at least on the parts relied on by appellants. If based on section 112, it is on the requirement thereof that "The specification shall contain a written description of the invention ***." (Emphasis ours.) We have a specification which describes appellants' invention. The issue here is in no wise a question of its compliance with section 112, it is a question of fact: Is the compound of claim 13 described therein? Does the specification convey clearly to those skilled in the art, to whom it is addressed, in any way, the information that appellants invented that specific compound? Having considered the specification in the light that has been shed on it by all the arguments pro and con, we conclude that it does not.  


[13.] Patent Act of 1952.

[14.] See, e.g., In re Ahlbrecht, 435 F.2d 908, 168 USPQ 293 (CCPA 1971); In re DiLeone, 436 F.2d 1404, 168 USPQ 592 (CCPA 1971); In re DiLeone, 436 F.2d 1033, 168 USPQ 598 (CCPA 1971); In re Lukach, 442 F.2d 967, 169 USPQ 795 (CCPA 1971); Fields v. Conover, 443 F.2d 1386, 170 USPQ 276 (CCPA 1971); In re Smith, 458 F.2d 1389, 173 USPQ 679 (CCPA 1972); and In re Welstead, 463 F.2d 1110, 174 USPQ 449 (CCPA 1972). See also the discussion of these cases and others of that vintage in Charles L. Gholz, Recent Developments in the CCPA Relating to the First Paragraph of 35 USC 112, 54 J. Pat. Off. Soc'y 768 (1972).

[15.] See, e.g., Vas-Cath, Inc. v. Mahurkar, 935 F.2d 1555, 1563, 19 USPQ2d 1111, 1116 (Fed. Cir. 1991), and cases cited therein.

[16.] See, e.g, Eiselstein v. Frank, 52 F.3d 1035, 1039, 34 USPQ 2d 1467, 1470 (Fed. Cir. 1995).

[17.] In re Smith and Hubin, 481 F.2d 901, 914, 178 USPQ 620, 623 (CCPA 1973), quoted in Vas-Cath, 935 F.2d at 1562, 19 USPQ2d at 1115-16.

[18.] In re Barker, 559 F.2d 588, 594, 194 USPQ 470, 475 (CCPA 1977), cert. denied, 434 U.S. 1064, 197 USPQ 271 (1978) (Markey, C.J., dissenting). See also Kevin S. Rhoades, The Section 112 'Description Requirement'--A Misbegotten Provision Confirmed, 74 J. Pat. & Trademark Off. Soc'y 869 (1992).

[19.] In critiquing the early and mid-1970s description requirement cases of the CCPA, one commentator stated that "the court's finest hours have not been spent in contemplation and resolution of the legal issues in this area of the law. Its decisions on the subject continue to confuse and confound." 2 Martin J. Adelman, Patent Law Perspectives §2.9[2.-3-1] (2d ed. 1997).

[20.] 541 F.2d 257, 191 USPQ 90 (CCPA 1976). As I read the case law, Wertheim marked a change toward a generally more liberal and common sense application of the law in description requirement cases, that reached its apogee in strict application in In re Welstead, 463 F.2d 1110, 174 USPQ 449 (CCPA 1972) . The CCPA itself acknowledged on at least three occasions that its cases of this era strictly applied the description requirement. See In re Arkley, 455 F.2d 586, 589, 172 USPQ 524, 527 (CCPA 1972); In re Smith, 458 F.2d 1389, 1394, 173 USPQ 679, 683 (CCPA 1972); and Wagoner v. Barger, 463 F.2d 1377, 1380, 175 USPQ 85, 87 (CCPA 1972). The board was subsequently scolded for taking the CCPA seriously. In re Edwards, 568 F.2d 1349, 1354, 196 USPQ 465, 469 (CCPA 1978).

[21.] 444 F.2d 599, 607. 170 USPQ 213, 220 (CCPA 1971).

[22.] 541 F.2d at 263, 191 USPQ at 97.

[23. ]558 F.2d 1008, 194 USPQ 187 (CCPA 1977).

[24.] Id. at 1013 n.12, 194 USPQ at 191 n.12.

[25.]Id. at 1019, 194 USPQ at 196. This language, although couched in enablement requirement terms, was seemingly intended to be in reference to the description requirement, as the next case discussed in the text shows.

[26.] 562 F.2d 1245, 1249-50, 195 USPQ 434, 438 (CCPA 1977).

[27.] Id. at 1250 (quoting Engineering Development Laboratories v. Radio Corp. of America, 153 F.2d 523, 526-27, 68 USPQ 238, 241-42 (2d Cir. 1946)).

[28.] 231 USPQ 393 (Bd. App. 1983), aff'd mem., 738 F.2d 453 (Fed. Cir. 1984).

[29.] The board cited In re Anderson, 471 F.2d 1237, 176 USPQ 331 (CCPA 1973). In Anderson, the court held that a claim to a wound dressing containing a layer carrying a medicant was described by disclosure in an original claim of identical terminology except for the term containing instead of carrying, combined with disclosure that the layer serves as a vehicle for medication, finding that the concept of "carrying" was described.

[30.] 231 USPQ at 394.

[31.] 231 USPQ 395 (Bd. App. 1983).

[32.] Since Grasselli involved an appeal to the Federal Circuit (Appeal No. 84-613), part of the prosecution history in the PTO, including the patent application as filed, is now a public record by virtue of its having been included in the appendix filed in that court pursuant to Fed. Cir. R. 30.

[33.] Unfortunately, the examiner's answer was not included in the appendix. See note 32, supra, and note 50, infra.

[34.] 231 USPQ at 394.

[35.] The PTO a number of years ago added a new chapter to the Manual on Patent Examining Procedure (MPEP)--Chapter 2100--entitled "Patentability," which purports to summarize the patent law. In §2173.05(i), titled "Negative Limitations," which falls under the broader §2173.05, titled "Specific Topics Related to Issues Under 35 USC 112, Second Paragraph," Grasselli is cited, but Johnson is not. Neither case is cited in §2163, titled "The Written Description Requirement," nor in any of the subsections thereunder. Indeed, Johnson is not cited anywhere in the MPEP. Not only does the PTO appear to be unaware of Johnson, the PTO reviewing tribunals appear to be unaware of it, too, since, except for Driscoll, no published decision of the CCPA or Federal Circuit appears to have cited Johnson for the proposition of law relied on here.

[36.] Rhoades, supra note 18, at 893-94.

[37.] 443 F.2d 1386, 170 USPQ 276 (CCPA 1971).

[38.] See cases cited supra note 14.

[39.] 443 F.2d at 1392, 170 USPQ at 280.

[40.] It is not clear what gloss the CCPA intended to put on the term "subservient" in the above-quoted passage in the text from Fields v. Conover. Nevertheless, the incentive to obtain any patent should not be diminished, not only those that would be dominated by a broader genus patent.

[41.] Thus, element M would be within the scope of Brown's claims. See In re Janakirama-Rao, 317 F.2d 951, 137 USPQ 893 (CCPA 1963), cited with approval in In re Herz, 537 F.2d 549, 551-52, 190 USPQ 461, 463 (CCPA 1976).

[42.] U.S. Constitution, Art. I, Sec. 8, Cl. 8.

[43.] An example of the difficulty of achieving a meeting of the minds in applying the possession test is illustrated in Ralston Purina Co. v. Far-Mar-Co. Inc., 772 F.2d 1570, 227 USPQ 177 (Fed. Cir. 1985). That case involved an issue of whether disclosure in a grandparent application provided description support for a variety of different and unrelated range limitations not literally disclosed in the patent. In a 2-1 decision, the Federal Circuit affirmed-in-part and reversed-in-part. In fact, there was a three way split between the district judge, the two-judge majority, and the dissenting judge, with the various factions lining up on different sides--sometimes together, sometimes opposed--depending on the particular issue.

[44.] The court stated:

Working backward from a knowledge of chlorpropamide, that is, by hindsight, it is all very clear what route one would travel through the forest of the specification to arrive at it. But looking at the problem, as we must, from the standpoint of one with no foreknowledge of the specific compound, it is our considered opinion that the board was correct . . . .

Ruschig, 379 F.2d at 995, 154 USPQ at 123 (emphasis added).

[45.] See Lockwood v. American Airlines Inc., 107 F.3d 1565, 1572, 41 USPQ2d 1961, 1966 (Fed. Cir. 1997).

[46.] See In re Piasecki, 745 F.2d 1468, 1471-73, 223 USPQ 785, 787-88 (Fed. Cir. 1984).

[47.] According to the views of the late Judge Nies of the Federal Circuit, Gray would also be denied relief under the doctrine of equivalents. See Roton Barrier, Inc. v. The Stanley Works, 79 F.3d 1112, 1128, 37 USPQ2d 1816, 1828 (Fed. Cir. 1996) (Nies, J., additional views) ("A substitution in a patented invention cannot be both non-obvious and insubstantial.").

[48.]See Fujikawa v. Wattanasin, 93 F.3d 1559, 39 USPQ 2d 1895 (Fed. Cir. 1996). In that case, the court resurrected and extended the 30-year old metaphor about forests, trees, and blazemarks conjured up in the first description requirement case, i.e., Ruschig. In an unusual twist, a party sought to add an additional subgenus count to an interference based on the disclosure of his opponent, while the opponent argued that his own disclosure did not describe the subgenus. The court agreed. It stated:

Many years ago our predecessor court graphically articulated this standard by analogizing a genus and its constituent species to a forest and its trees.

Id. at 1570, 39 USPQ2d at 1905. The court later continued:

Were we to extend Ruschig's metaphor to this case, we would say that it is easy to bypass a tree in the forest, even one that lies close to the trail, unless the point at which one must leave the trail to find the tree is well marked. Wattanasin's preferred embodiments do blaze a trail through the forest; one that runs close by Fujikawa's proposed tree. His application, however, does not direct one to the proposed tree in particular, and does not teach the point at which one would leave the trail to find it.

Id. at 1571.

[49.] 30 USPQ2d 1234 (Bd. Pat. App. & Inter. 1993). See also In re Wright, 866 F.2d 422, 9 USPQ2d 1649 (Fed. Cir. 1989) (term "not permanently fixed" not literally disclosed held to be described by unequivocal teaching of absence of permanently fixed microcapsules).

[50.] 30 USPQ2d at 1236. Unless the examiner's answer dropped the rejection regarding the Jepson preamble negative limitations from the final rejection, which is doubtful, in Grasselli, the board's summary thereof in Parks is incomplete, and thus, incorrect.

[51.] Id.

[52.] Id.

[53.] 76 F.3d 1168, 37 USPQ2d 1578 (Fed. Cir. 1996).

[54.] One prescient commentator (and colleague), some 25 years ago in the midst of the early, strictly-applied description requirement cases (see, e.g. cases cited supra note 14), and reminiscent of the comment by Judge Learned Hand quoted previously in the text (see supra note 27), foresaw a "danger" that could "lead to rampant, socially dysfunctional 'paper chemistry' and the equivalent in other arts. With the present capability of business machines, it is relatively simple to go from a structural formula reading on tens of thousands of chemical compounds to a printout naming the compounds individually. If the generic element in question is a temperature or pressure range, a range in ingredient proportions, or the like, it is even easier to recite mechanically legions of subgeneric descriptions drawn to narrower ranges or specific values within the generic range." Gholz, supra note 14, at 787-88. This has happened to some extent, I believe, at least in some of the organic chemical arts. How much of it is "socially dysfunctional" is left to be determined by the sociologists and the economists. It will, however, alleviate the problems discussed herein, at least until a better solution is found, as I have tried to advocate in this paper.