By: Benjamin J. Barish, Adv., B.Sc. in E.E., 1946, Illinois Institute of Technology, Chicago;

J.D., 1950, LL.M. 1951, George Washington University, Washington, D.C.

(All opinions expressed herein are those solely of the author)

A woman went to the market to buy a chicken. She approached a stand covered with a pile of chickens for sale, grasped a chicken, held it up to the light, spread the wings and legs, one way, then the other, and carefully examined, probed and sniffed each part of the chicken, but eventually discarded the chicken and picked up another one for examination. After she had thus examined and discarded a number of chickens, the shopkeeper, witnessing all this with patient silence, could hold back no longer: “Lady”, he asked, “could you yourself pass such an examination?”

How many of the issued patents in the field of high technology could confidently pass such an examination with respect to the best mode requirement of 35 USC 112, first paragraph, as interpreted by recent court decisions, particularly when the invention relates to a highly technical development, and the patent application was filed just before commerica1 production? A review of recent decisions on this issue, as set forth below, indicates that, at best, the answer to this question is unpredictable, and in many or most cases would be negative because the tests applied are both subjective and unrealistic in today’s high technology world.

The Present Situation

Take the case of a highly technical development for an engine control system that you have been requested to protect by one or more patent applications. The engine control system includes the following main components: (a) a fuel feeding system for feeding fuel to the engine; (b) a plurality of sensors for sensing predetermined conditions in the engine; (c) an electrical processor for processing the outputs of the sensors according to certain algorithms to output control signals; and (d) control devices, such as actuators, valves, etc., for controlling the fuel to the engine in accordance with the output control signals.

Assume there is novelty in certain aspects of the overall system, as well as in one or more of the above components, so that it may be necessary to file more than one patent application to cover all the novel features. However, only one system has been developed and optimized for production. Each of the above four components involves considerable detailed information including information characterised as manufacturing data, customer requirements, and trade secrets. Such information may include, for example, dimensions, tolerances, materials, heat treatments, performance and design specifications, assembly methods, algorithms of processor operations, commercial name of outside-supplied components, etc. A written description of all such information of the optimized system developed for commercial production would require hundreds or thousands of pages or documents.

It is clear that such detailed information involved a tremendous investment on the part of the developer. Such information might be, and frequently is, included in a separate “know-how package” if the optimized form of the development is licensed to another for production.

The attorney charged with drafting the patent application or applications is first faced with the dilemma of determining exactly what information to include in the patent application to assure compliance with the best mode requirement, as well as the enabling requirement, of 35 USC 112, first paragraph. Since only one mode has been developed and optimized, the safe course of action would be to include in the patent application(s) all the information including the detailed manufacturing data, customer requirements and trade secrets embodied in the many hundreds or thousands of documents, particularly in view of the holding in the leading case on this issue, Chemcast Corp. v. Arco Industries Corp. 913 F.2d 923 (Fed. Cir.. 1990), at 930:

“–Whether characterizable as “manufacturing data” “customer requirements,” or even “trade secrets,” information necessary to practice the best mode simply must be disclosed”.

Not only would this be impractical (or perhaps even impossible), but an attorney attempting to do this, in all probability, would be performing a disservice to the client, to the Patent Office, and to the public. Thus, the eventual publication of this valuable know-how information to possible competitors could seriously compromise the client’s investment in developing this information. Including this detailed information in the patent application would require the Patent Office to accommodate this great bulk of material in its files and the Examiner to consider it in the examination for novelty, thereby obscuring the really novel features sought to be patented. It would have little value to the public except to competitors who wished to manufacture the optimized system in all its details. Even then, the value of such information would be very limited since the system, if successfully patented, could not be legally manufactured without license until the patent expires; and by that time, much of the information would probably have become outdated.

If anything less than all the technical details, requiring hundreds or thousands of pages or documents involved in the optimized development, is included in the patent application, the attorney runs the risk of omitting something which may later be considered to be a feature of the patented development sufficiently significant to risk the patent being invalidated on the grounds of failure to comply with the best mode requirement.

In any event, this requirement, as set forth in the above quotation from the Chemcast decision, effectively invites everyone charged with infringing a patent to conduct an extensive “sniff and probe” examination in an attempt to identify any technical information which was known to the inventor or employer at the time the patent application was filed, which was not disclosed in the patent application, and which had some significant effect on the optimized implementation of the invention, in order to invalidate the patent for failure to comply with the best mode requirment. Even if all the details in these hundreds or thousands of documents are disclosed in the patent, there is still no certainty that the patent will be immune from such attack because of the impossibility of assuring that all relevant information known to all the individuals involved in the optimized development was properly considered or documented, particularly in a sophisticated invention in high technology.

The above Chemcast decision applied a two-step analysis for the “best mode requirement”: The first step is to determine whether the inventor, at the time the patent application was filed, knew of a mode of practicing the claimed invention considered by the inventor to be better than any other. If the inventor in fact contemplated such a preferred mode, the second step of the analysis is to compare what the inventor knew with what the inventor disclosed, and particularly to determine whether all such information “necessary to practice the best mode” was indeed disclosed.

The problem with the Chemcast two-step test is that it “begs the issue”.
Thus, the first inquiry – “whether the inventor in fact contemplated such a preferred mode” – presumably should always be answered in the affirmative since 35 USC 112 requires that the inventor:

“shall set forth the best mode contemplated by the inventor of carrying out his invention”.

It is hardly conceivable that an inventor can avoid this issue by merely refusing to admit that a best mode was indeed contemplated before the application was filed.

Therefore, we are left only with the second inquiry; but this second inquiry merely restates the ultimate issue involved in determining compliance or non-compliance with the best mode requirement, and provides no guidance, except in the all-inclusive statement quoted above, for making this determination.

Subsequent court decisions have variously applied the above two-step Chemcast test, and/or a “concealment” test, among many other tests. Some cases held that to invalidate the patent “concealment” must be based on knowledge coupled with a specific intent to conceal, whereas other cases held that no specific intent is necessary. However, the problem with using “concealment” as a test is that anything known but not disclosed may be considered as “concealed”. Also, since every detail of the manufacturing data included in the hundreds or thousands of documents describing a highly-technical optimized development can be considered as having some affect on the manufacture or operation of the development, a “concealment” test is not only unrealistic, but is highly subjective, depending on the sympathies and inclinations of the decider (judge or jury), and produces unpredictable results. It also encourages one to “sniff and probe” without limitation to locate something that can be argued as concealed.

This dilemma presented by recent court decisions regarding the “best mode requirement” is well highlighted in an exhaustive analysis of the current situation by Stephen B. Walmsley; Best Mode: A Plea to Repair or Sacrifice This Broken Requirement of United States Patent Law; Walmsley12-9 Mich.Telecomm.Tech.L. Rev. 125 (2002). In this article, the author traced the evolution of this requirement and discussed the multitude of standards set forth in the cited cases for assessing the disclosure needed to comply with this requirement. The various standards were summarized into the following seven categories: (1) the “claims-only” standard; (2) the “essentiality” standard; (3) the “essentiality and novelty” standard; (4) the “necessity” standard; (5) the “necessary and nonobvious” standard; (6) the “material effect” standard; and (7) the “relation” standard. A diagram was attached (Diagram 2 of the Appendix) setting forth these seven standards, including the cases on which each is based. The article concludes with the following Summary (page 167):

“The definition of the best mode requirement of U.S. patent law is unclear and the scope unstable. Instability in the scope of the best mode requirement leads to great expenses incurred in over-complying with the statute and to even greater risks in under-complying with the statute. Such unnecessary expenses and risks are a great disservice to the patent system. Therefore, the best mode requirement desperately needs to be stabilized by an en banc ruling of the CAFC. In the alternative, the best mode requirement should be beneficially sacrificed as a bargaining chip during negotiations with other countries in furtherance of international patent law harmonization efforts. In the meantime, however, inventors are advised to disclose everything important enough to have been documented during the project development phase of their invention process.”

This dilemma, as to the information to be included by the US practitioner in a patent application in order to meet the best mode requirement, is no less for the foreign practitioner when preparing the first patent application to be filed in the foreign country, since the foreign application must also meet all the requirements of 35 USC 112, first paragraph, if it is to support a claim for its priority date under the International Convention.

Clearly, therefore, there is an urgent need for a realistic, objective test for the “best mode requirement”. Before formulating such a test, it is first necessary to consider the legislative intent in 35 USC 112, first paragraph.

The Legislative Intent

The first paragraph of 35 USC 112 provides:

“The specification shall 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 any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.”

The “enabling” requirement in the above-quoted passage appears first and involves primarily a quantitative test, namely sufficient information “to enable” an implementation of invention. The “best mode” requirement follows the “enabling” requirement in the above passage and appears to involve primarily a qualitative test qualifying the “enabling” requirement, namely by specifying the quality or type of mode to be described to meet the “enabling” requirement.

It is believed this indeed was the legislative intent in 35 USC 112, first paragraph. That is, it is not believed that the legislative intent was that the Applicant set forth all the information that may be involved in the “best mode contemplated by the inventor”, including all the know-how, technical details, etc. Rather, it is believed the legislative intent was that the Applicant set forth enough information (quantitatively speaking) to meet the “enabling requirement”, and that such information (qualitatively speaking) relate to “the best mode contemplated” by the inventor at the time the patent application was filed, rather than to an inferior mode. In short, it is believed the legislative intent was:

(a) in the enabling requirement, to require that the applicant teach the public sufficient information to enable the practice of the invention (primarily a quantitative test), and

(b) in the best mode requirement (which follows the enabling requirement) to require that the information set forth to meet the enabling requirement relate to the best mode, and not to an inferior mode (primarily a qualitative test).

While issues concerning the “enabling requirement” arose from the start and were resolved without particular difficulty, it was only after Judge Rich’s opinion in In re Nelson (126 USPQ 242, CCPA 1960) that difficult issues began to arise with respect to the best mode requirement, because of the statement:

“Hence Section 112 calls for description in ‘full, clear, concise and exact terms’ and the ‘best mode’ requirement does not permit an inventor to disclose only what he knows to be his second-best embodiment, retaining the best for himself”.

However, Judge Rich’s opinion in the Nelson case itself suggests that the “best mode” requirement concerns primarily a qualitative test and not a quantitative test: namely, that the information disclosed to meet the enabling requirement must be directed, not to a second-best embodiment, but rather to the best embodiment known to the inventor at the time the application is filed.

The Nelson decision was followed by decisions in other courts applying a “concealment” test. Some decisions required both knowledge and special intent to conceal a best mode in order to invalidate on these grounds (Benger Laboratories. Ltd. v. R.K. Laros Co. 133 USPQ 693, third Cir. 1963); Carter-Wallace, Inc. v. Riverton Laboratories, Inc., 167 USPQ 656 (second Cir. 1970); whereas other cases held that specific intent to conceal is not necessary (Dale Electronics. Inc, v. R.C.L. Electronics. Inc. 180 USPQ 225, first Cir. It is believed these cases wrongly applied a primarily “quantitative” test, namely that all the information should be supplied regarding the best mode, rather than what is believed to be primarily a qualitative test in 35 USC 112, first paragraph, and in the Nelson case.

The “quantitative” test for the best mode requirement was particularly applied in Union Carbide Corp. v. Borq-Water Corp. 193 USPQ 1 (sixth Cir. 1977). This decision involved a patent directed to a process for moulding foamed thermoplastic products, and held that the inventor’s failure to disclose a particular valve and a particular extruder used to perform the process violated the best mode requirement, even through the new valve was the subject of a subsequently filed U.S. patent application. The Court so held despite the fact that the patentee had signed up 54 licensees, each of which had paid for a “know-how package” containing extensive written specifications, including the specially designed extruder.

However, it is not believed the legislative intent, in 35 USC 112, first paragraph, was to require the patent application also to include clearly know-how information relating to an optimized form of the best mode of the invention. If this indeed had been the legislative intent, this would have effectively precluded inventors from protecting “know-how” and trade secrets with respect to patented inventions, as always permitted and as now quite common.

Also relevant to the issue of the “legislative intent” in 35 USC 112, first paragraph, is the scope of the relevant subject matter required to be described in the specification. Various courts on various occasions have applied many conflicting standards as to the scope of such subject matter, summarized in the above-cited Walmsley article into seven categories. It is believed, however, that the legislative intent with respect to the scope of the relevant subject matter is revealed by the actual words appearing in the first and second paragraphs of 35 USC 112. Thus, the first paragraph states:

“The specification shall contain a written description of the invention, …”

and the second paragraph of 35 USC 112 states:

“The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. …”

It is believed that when both sentences are properly construed together, the legislative intent in defining the required description of the invention in the first paragraph refers to the invention defined in the claims specified in the second paragraph.

Determining the legislative intent with respect to the “best mode requirement” of 35 USC 112 should also take into consideration the legislative intent of other provisions of the same statute. For example, 35 USC 181 – 188 deal with “Secrecy of Certain Inventions and Filing Patent Applications in Foreign Countries”. These sections of the U. S. Patent Law have the purpose of preventing publication of information which may be detrimental to national security. Thus, the legislative intent of 35 USC 112, to require a full disclosure, should be reconciled with the legislative intent of 35 USC 181, et seq., to prevent public disclosure of information which may be detrimental to national security.

For example, a development may be made having both a military application and a commercial application. The preferred embodiment of the invention (the best mode) contemplated by the inventor may be the military application whose publication would be barred under the above national security provisions of 35 USC 181, et seq. Therefore if that best mode is indeed described in the specification as required under
35 USC 112, a “Secrecy Order” would be applied under 35 USC 181, et seq. barring publication of the invention. Such “a Secrecy Order” would thus prevent the applicant from exploiting the invention, would deprive the public of the information regarding the invention, and would subject the Government to a claim for reasonable compensation (35 USC 183).

On the other hand, if publication of the commercial, inferior embodiment would not be detrimental to national security, it would appear preferable, and consistent with the “legislative intent” with respect to the above two conflicting statutory provisions, to permit the inventor to describe the inferior (commercial) embodiment, rather than the preferred (military) embodiment. Such a legislative intent would benefit both the public and the inventor and would thus best implement the Constitutional basis for the patent laws (Article 1, Sec. 8):

“– to promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries”.

Such an approach would be consistent with the approach already taken by the courts in an analogous situation, illustrated by the “fun example” described in the Walmsley article. In this example, an inventor discovered that acid deposits from terminal posts of a car battery could be removed by using a carbonated soft drink, and that best results were obtained when using “Coca-Cola” (Registered Trade Mark of the Coca-Cola Company). Since the composition of a product identified by a trade mark may be kept secret (as indeed it is in this case), and may be changed at will by the manufacturer, the best mode disclosure requirement of 35 USC 112, first paragraph, if interpreted strictly and absolutely, would require a disclosure of the actual composition of “Coca-Cola”. However, as pointed out in the Walmsley article, it has been held that the disclosure of the trade name but not the formula, of a cleaning solution used in the claimed invention was sufficient for best mode purposes (see Randomex, Inc. v Scopus Corp., 849 f.2d 585 (Fed Cir. 1988)).

Summarizing the foregoing, it is believed that the legislative intent in
35 USC 112, first paragraph, was:

(a) to apply primarily a quantitative test to the “enabling requirement”, i.e., to require sufficient information to enable a person skilled in the art to practice the invention;

(b) to apply primarily a qualitative test to the “best mode” requirement, i.e., to require that the information supplied in the specification be directed, or at least include, the “best mode” contemplated by the inventor for carrying out the invention;

(c) to apply a claims standard in determining the scope of the subject matter to be disclosed, i.e., to measure the information required to be supplied in the specification by the scope of the invention as defined in the claims; and

(d) to apply a public policy test to the “best mode” requirement, i.e., to permit an inferior mode to serve as the best mode where the actual best mode may not be legally permissible or practically possible, e.g., because of national security considerations, trade secret considerations, etc.

It is to be emphasized that the above tests apply only to the “best mode requirement” of 35 USC 112, first paragraph, and not to the “enabling requirement”. The “enabling requirement”, it is believed, was intended to be an independent requirement, separate and distinct from the “best mode” requirement, in determining compliance or non-compliance with 35 USC 112, first paragraph.

Proposed Four-Prong Approach

In the light of the foregoing analysis, a four-prong approach, believed to be realistic, objective and consistent with the legislative intent, is proposed below to test the specification disclosure with compliance with 35 USC 112, first paragraph:

1. Determine whether the specification contains sufficient information to meet the enabling requirement (primarily a quantitative test).

2. If so, determine whether the disclosed information relates to the “best mode” implementation of the invention, or is it restricted to an inferior mode (primarily a qualitative test). If it does not relate to or include the “best mode”, then presumably there was an intent to conceal the best mode and to lead the public to an inferior mode, and therefore the best mode requirement is presumably violated.

3. In testing both the enabling requirement and the best mode requirement, consider that the scope of the subject matter required to be disclosed is that relating to the invention as defined in the claims. For example, if the claims recite a fuel feeding system including valves, controls, actuators, etc., and no novelty is claimed in the construction of the valves, controls, or actuators, it presumably would not be essential to disclose the details of such valves, controls, or actuators to meet the best mode requirement, assuming of course that the enabling requirement is met.

4. If the disclosed embodiment does not include or relate to the best mode, determine whether there is a legal justification, based on public policy, for not including a description of the best mode. As indicated above, one possible legal justification would be where the best mode involves a national security embodiment which may not be publicly disclosed and which, if described in the patent application, would subject the application to a Secrecy Order. As discussed above, public policy should justify permitting the disclosure of the best non-security implementation even though it was not the actual best mode. Another possible justification is illustrated by the “Coca Cola” example mentioned above, wherein the courts have already held that, in certain cases, failure to disclose a secret formulation identified by a trademark does not necessarily violate 35 USC 112, first paragraph.

By thus applying, per prongs 1 and 2 above, a primarily quantitative test to the enabling requirement, and a primarily qualitative test to the best mode requirement, Section 35 USC 112, first paragraph, would require: (1) that the description provide sufficient information to meet the enabling requirement, and (2) that the such supplied information relate to the best mode requirement. In other words, this provision would require that the specification include a description of the best mode, and that such a description be sufficient to meet the enabling requirement, but need not include all the know-how, trade-secret, etc., information known to the inventor for “optimizing” the best mode.

By applying, per prong 3 above, a “claims standard” to the scope of the subject matter required to be included in the description, numerous technical details not related to the claimed novelty would not be required to be included in the description, thereby relieving the inventor from the burden, and possible commercial implications, of disclosing technical details not directly related to the invention.
By applying, per prong 4 above, a “public policy” test permitting a “second best” mode in certain cases (e.g., because of a possible detriment to the national security), the constitutional purpose of the patent system, in promoting the progress of science and the useful arts, is best implemented, by making available to the public a description of the invention without compromising national security, while at the same time securing for limited times to the inventors the exclusive rights to their discoveries.

Replacing the present unrealistic, subjective tests for the best mode requirement as applied in recent decisions, by the above four-prong, realistic objective test, will provide a clear and stable definition of this requirement consistent with the apparent legislative intent. As a consequence, it will avoid, to a large extent, the many disadvantages in over-complying with this requirement, the risks in under-complying with this requirement, and last, but not least, the need to make the exhaustive “sniff and probe” type examination involved in the anecdote at the beginning of this article.

Benjamin J. Barish

Advocate, Patent Attorney

Ehrlich & Fenster

News & publications:
More Articles and Publications: