given them respect appropriate to their status. 236 On the other hand, when the facts before the court are significantly different, a new determination is being made-and the Patent Office decision may be completely honored and yet a different ultimate conclusion reached as to the question of invention. In such instances there is no reason to regard the court decisions as contrary to accepted doctrines of administrative finality, for the logical justification for such finality is largely destroyed.

One may well ask whether it is really necessary to have a patent system wherein the opportunity exists for a decision on one set of facts in the Patent Office and another set of facts in court. The ans

wer lies in experience that has shown that very frequently the Patent Office does not locate references that later become important in court; commercial success and like factors come to bear heavily on the issue of invention and cannot be considered in full perspective in the Patent Office; and prior use and like evidence inherently is not usually available in the Patent Office. It is just as unsound today to think in terms of foreclosing judicial inquiry on the invention question as it is to think in terms of repeating the unhappy experience of the Patent Act of 1793, when Patent Office consideration of this question was foreclosed.

A different question arises with respect to the review of district court decisions on the invention issue. Here the facts before the appellate court coincide with those upon which the lower court decision is based. Whatever may be said of the difficulties of reaching the decision on the invention question in the district court, it is certainly a fact decision of the kind that should be treated as such on review and not overturned unless "clearly erroneous."

Finally, in all consideration of the invention question the basic nature and purpose of the requirement should not be overlooked. The extent of changes over the prior art vary from exceedingly rare giant strides forward, to the common day-to-day engineering improvement. A patent system based on patents solely to the giant strides would fail of its purpose for the prospect of patent protection would be too remote to stimulate research and development effort and investment. On the other hand patents extending to simple engineering improvement would result in a hopeless multiplicity of patent rights and the system would serve to deter rather than encourage activity. Between these extremes there is a proper place for patent protection. Again, the emphasis must be upon individual fact situa

236 See Frost, Patent Office Performance in Perspective, 54 Mich. L. Rev. 591 (1956).

The interference decisions form a particularly interesting illustration of respect for Patent Office decisions. Here the initial decision is based on an inter partes proceeding in the Patent Office and there are usually no new issues and no significant new evidence when the matter of priority arises in a subsequent case on direct review of the Patent Office under title 35 United States Code, section 141 or section 146, or in an infringement suit where the defense is that the losing party in the interference was really the first inventor (35 U. S. C. 102 (g)). The Supreme Court early laid down the rule that the Patent Office decision must stand in any subsequent suit between the same parties "unless the contrary is established by testimony which in character and amount carries thorough conviction" (Morgan v. Daniels, 153 U. S. 120, 125 (1894)). The Court held in Radio Corporation of America v. Radio Eng. Laboratories, footnote 235, supra, that this rule applies in considerable measure when the litigants are strangers to the proceeding in the Patent Office. Considering the various expressions as to the presumption of validity of the patent the Court stated: "*** Through all the verbal variances, however, there runs this common core of thought and truth, that one otherwise an infringer who assails the validity of a patent fair upon its face bears a heavy burden of persuasion, and fails unless his evidence has more than a dubious preponderance" (293 U. S. at p. 8). It is of interest that in the Radio Corporation case the Patent Office had awarded priority to Armstrong and that it was the reversing decision of the Court of Appeals for the District of Columbia in favor of DeForest (298 Fed. 1006) that gave rise to issuance of the patent and hence was the decision the Court was relying upon.

For a recent decision squarely refusing to follow the Patent Office on the art before the Patent Office see Wilson Athletic Goods v. Kennedy, 233 F. 2d 280 (2d Cir. 1956). See also, pp. 176-185 and 287-293, hearings, October 10-12, 1955, pursuant to S. Res. 92 (reporting the results of a study by Mr. P. J. Federico of the Patent Office of 50 recent cases where patents were held invalid by courts of appeals, and showing that of the 40 cases where the patents were held invalid on prior art only six involved such holdings on the same ** as was before the Patent Office).

tions and the skill of the art test-coupled with the thought that the opportunity for patent protection should be sufficiently real to give rise to the research and development competition that lies at the heart of the whole system.

Much has been said of the "standard of invention" and the high level assertedly demanded by recent Supreme Court decisions.237 Indeed, one segment of the Supreme Court takes the position that the Constitution itself embodies a particular and very severe standard in this connection.238 Yet, on their facts, it is at least arguable that the Supreme Court decisions really impose no new standards,239 and statistics may be assembled to show that current decisions are no less favorable to the Patent Office determination than those of an earlier day 240

237 See e. g., cases cited at footnote 199, supra.

238 The standard of patentability is a constitutional standard; and the question of validity of a patent is a question of law.*** The Court fashioned in Graver Mfg. Co. v. Linde Co., 336 U. S. 271, 275, a rule for patent cases to the effect that this Court will not disturb a finding of invention made by two lower courts, in the absence of a very obvious and exceptional showing of error. That rule, imported from other fields, never had a place in patent law. Having served its purpose in Graver Mfg. Co. v. Linde Co., it is now in substance rejected. The Court now recognizes what has long been apparent in our cases: that it is the 'standard of invention' that controls. That is present in every case where the validity of a patent is in issue. It is that question which the Court must decide. No 'finding of fact' can be a substitute for it in any case. The question of invention goes back to the constitutional standard in every case. We speak with final authority on that constitutional issue as we do on many others." Mr. Justice Douglas, concurring in Great Atlantic and Pacific Tea Co. v. Supermarket Equip. Corp., 340 U. S. 147, 156 (1950). Cf. Alfred Bell &Co. Ltd. v. Catalda Fine Arts, Inc., footnote 227, supra.

239 The two decisions most frequently referred to as indicating a new standard of invention are Cuno Eng. Corp. v. The Automatic Devices Corp., 314 U. S. 84 (1941), and Great A. & P. Tea Co. v. Supermarket Equip. Corp., 340 U. S. 147 (1950).

In the Cuno case counsel argued that the "skill of the art" test laid down in Hotchkiss v. Greenwood, 11 How. (52 U. S. 248 (1850)), was wrong and that "degree of invention or ingenuity is not a test contemplated by the Constitution and the patent laws to determine whether or not an invention or discovery shall receive protection ***." See footnote 197, supra. The sweeping statements of the opinion appear to be more in the nature of an emphatic answer to this contention than an effort to establish a new standard. Indeed, the last sentence of the opinion states that the considerations set forth in the opinion "prevent any relaxation of the rule of the Hotchkiss case," (314 U. S. at p. 92). As the concurring opinion points out, the infringement charged was quite different in construction than the showing of the patent. It should also be noted that the case came up after two circuit courts of appeal had reached conflicting opinions on validity.

In the A. & P. case the structure of the patent was the three sided bottomless frame used in supermarket checking counters. The device had enjoyed great commercial success. Nonetheless, it is in that class of articles that are of obvious simplicity and tend to evoke violent reactions-one way or the other-as to the presence of invention. In any event, the district court and the court of appeals had both concluded that "the bottomless tray is not novel," and had sustained the patent on the ground that the extension provided on the counter to receive the three-sided rack was "decidedly a novel feature" (78 F. Supp. 391) Claim 4 in suit did not refer to the extension. Claims 5 and 6 arguably referred to it, but a majority of the Supreme Court considered otherwise. Reversal followed-on the theory that what had been relied upon below to sustain the patent (the extension) was not in the patent claims. The Court added that merely increasing the length of the counter could not of itself be invention. The major discussion of the majority opinion, however, was directed to the requirement that the elements of a combination must "cooperate" to produce something unusual-a point not considered by the lower courts. The concurring opinion of Mr. Justice Douglas, famous for its sweeping criticism of the Patent Office, was joined in only by Mr. Justice Black.

240 The following tabulation shows the trends:

Figures for the 1926-54 period are taken from p. 183, hearings, October 10-12, 1955, pursuant to S. Res. 92. Other data are taken from tabulations prepared on a generally similar basis. Due to the effects of the institution of circuit courts of appeal in 1895, the introduction of discretionary certiorari jurisdiction in patent cases in 1926, and the statistically insignificant number of cases in some periods, the above data are

While it is thus possible to discount the "new doctrinal trend," the fact remains that the sweeping language used in some of the recent Supreme Court decisions reveals in at least some measure an attempt to impose more rigorous standards upon the Patent Office and the lower courts. The statutory expression of section 103 of the 1952 Patent Code should aid in eliminating the confusion that has been generated in this fashion. The recent opinion of Judge Learned Hand in Lyon v. Bausch and Lomb 241 is noteworthy in this connection because it squarely discards the statements of the recent Supreme Court decisions as the measure of invention. While there may yet be some occasion for additional legislative declaration of policy or an amendment to the statutes, the matter is now in a state of flux offering substantial prospect that the problem will be worked out within the confines of the present statutes.

The above comments do not point to any new formula by which the invention question may be resolved. They do not even provide a hope for uniformity between invention decisions in different cases. They do, however, suggest that the emphasis ought to be placed on the factual nature of each decision in question-and that it be given a degree of finality in accord with the extent the factual basis is the same in future decision. The Patent Office decision is not and should not be final-but on the facts before the Office it certainly deserves great weight. The district court decision should be treated as a fact decision and so honored on review. And above all, there should be no confusion between what is essentially fact determination and what is law. Finally-cutting across all of these concepts-the courts in many cases can find considerable light on the nature of the invention question by comparing the position of the patentee as to the prior art with that respecting asserted infringement and insisting that consistent standards be applied.

One general consideration deserves final emphasis. It should not be assumed that some overriding public purpose is served every time an excuse is found to overturn a patent. In nearly every instance the real question is not whether the defendant shall stay in business but rather whether it can make the particular structure in question or use the particular process of the patent. Experience with the decisions such as the Ray-O-Vac case discussed above-shows that a decision of patent validity is more likely to lead to desirable research effort than otherwise. More often than not such effort leads to improvements superior to that which has been patented. There may be endless points of view on the question of whether a specific change exceeds the "skill of the art"-but there should be little doubt that if the change is of the kind that a trier of fact reasonably can classify as inventive the public interest is served more by honoring the holding made than striking it down because of the doubt that is rarely absent in litigated cases.

B. THE PATENT OFFICE

The Patent Office occupies a crucial position in the operation of the patent system, for it is charged with the task of initially passing upon

not truly comparable. Nevertheless, they do show that current experience is not such a departure from the past as is sometimes suggested.

Some commentators have found that the decisions really display only a trend toward earlier and stricter views on patentability. See, e. g., Smith, Recent Developments in Patent Law, 44 Mich. L. Rev. 899 (1946). Compare Davis, The Impact of Recent Supreme Court Cases on the Question of Patentable Inventions, 44 Ill. L. Rev. 41 (1949).

241 224 F.2d 530, 543–7 (2 Čir. 1955).

applications for letters patent to determine whether a patent should issue and to fix the scope of the patent, if issued. The unfortunate experience from 1793 to 1836 amply demonstrates that patents must be subject to examination before issuance. Current experience additionally shows that the interests of all parties, as well as the maintenance of research and development competition, demand that the Patent Office be reasonably current with its work and in step with the courts on the issues it decides.

For some time the Patent Office has been fighting a losing battle with the influx of new patent applications and has been falling behind on the equally important task of effectively classifying the patents and other items in the search files. An 8-year program has been established to work off the backlog and bring operations to a current basis.242

It is of crucial importance that the Patent Office receive appropriations adequate to its task. More than private interests are involved. As a storehouse of technical knowledge the Office serves all industry. However, the patents and other items are so numerous that an adequate classification system is of paramount importance. Today due to inadequate budgets over a long period-the classification system is out of date. This same classification system is necessary to serve the patent examiners making searches on patent applications. Here the inadequacies of classification turn up in the form of a failure to cite the best prior art.

The Patent Office budget should also be adequate to assure prompt and complete patent examination. Every consideration demands. that patents issue as promptly as possible after filing. In many instances the patent disclosure is the first teaching of a new development to the trade. Prompt patent issuance serves to trigger competitive effort based on the patent and thus speeds the all-important pace of competition in technical development. An adequate budget should additionally free the patent examiners from an unreasonable case load and thereby permit more thorough and effective search effort before patent issuance. We do not know what the patent mortality in the courts will be when the Patent Office is current in its work, the classification problem is overcome, and the workload on the examiners is reduced from the present level. We do know, however, that at the present time the great bulk of decisions of patent invalidity are based on art not considered by the patent examiner.243 There is reason to predict that improved Patent Office search activity will lead to a more favorable court record. Additionally, it may well turn out that the collection and citation of more complete prior art in the Patent Office, coupled with a less pressing work load, will lead to more precise patent claims and file wrapper arguments serving to improve the court experience with patents to an extent greater than analysis based on the patent invalidity decisions might indicate.

The Congress has recently passed a substantially increased Patent Office appropriation. This is a step in the right direction. While it will take time for the backlog of the past to be cut down and the effects of the increased budget to be felt in full, it is nevertheless most encouraging that the trend is in the direction of improvement.

242 Hearings, October 10-12, 1955, pursuant to S. Res. 92, pp. 162, 198.

243 Id. at p. 183.

1. Reducing the Patent Office load at its source-defensive patent appli

cations

There is also an opportunity to relieve the Patent Office load at its source. Currently many patent applications are filed for "defensive" purposes to minimize the risk of patent infringement charges based on later filed applications. Most, if not all, Government patent applications fall in this category.244 Industry likewise files many patent applications of this kind.25 The number of such applications has been estimated at one-third the total of all patent applications.246 Whether this figure is high or low is not important-for by any standard the number is substantial and they necessarily interfere with Patent Office work on other applications.

The odd fact is that the law now places such a premium upon patent applications as a defensive measure that common prudence dictates their use whenever a manufacturer or the Government develops and intends to use an arguably patentable product or process. At the application stage the defensive application is ordinarily thrown into interference with any adverse patent application which, if issued, would give rise to infringement charges.247 The issue of priority can then be resolved at the Patent Office level-with the full benefit of a "constructive reduction to practice" as of the filing date,248 before any question of damages has arisen,249 and in a procedural atmosphere generally more favorable than court proceedings.250 If there is no interference it is still to the advantage of the applicant to make every effort to persuade the Patent Office to issue the defensive application, for in that event the issued patent relates back to become a constructive reduction to practice as of its filing date and with respect to all that it discloses.251 It thus gives defensive assurance that is avail

able in no other way.

The result is that we have a great body of patent applications filed for defense only-applications which the patent examiners must study for technical sufficiency even though no rights turning on such sufficiency are really sought; claims for which the patent examiners must make regular prior art searches even though the applicant has no desire to enforce the claims, if allowed; applications demanding the preparation of office actions rejecting or allowing the claims; and

244 See Government-Owned Patents and Inventions of Government Employees and Contractors, Second Report of the National Patent Planning Commission, 27 J. P. O. S. 76, 78 (1945). ("The main, if not the sole, purpose of any proprietary interest of the Government in patents has been one of protection against interference with the performance of governmental functions.") See also, Department of Justice, Investigation of Government Patent Practices and Policies, Vol. 1, pp. 126-131 (1947) and Kear v. Roder, 115 F. 2d 810 (C. C. P. A. 1940) (publication of invention made by Bureau of Standards employee not a reduction to practice).

245 See e.g., McHugh, Bell System Patents and Patent Licensing, Bell Telephone Magazine, January 1949. ("It is the practice to apply for United States patents upon the more important of these inventions so that the Bell System's right to use its own inventions in furnishing communications service may receive the assurance provided by the patent laws.") And see pt. 2, TNEC hearings, p. 256 (Ford Motor Co.), pp. 363-364 and 692 (General Motors Corp.).

246 Davis, Proposed Modifications in he Patent System, 12 Law and Contemp. Prob. 796, 799-800 (1947). 247 It is the practice of the Patent Office to conduct an interference search after an application is found to contain allowable subject matter. Such search systematically covers the other pending applications with the objective of avoiding issuance of an application containing matter interfering with another application. See sec. 1101.01 (c), Manual of Patent Examining Procedure. In addition, the owner of an application on file can copy the claim of an issued patent for interference purposes at any time within 1 year of issuance of the patent. See 35 U. S. C. 135.

248 Footnote 180, supra.

249 The patent right to exclude-and the opportunity to recover damages-begins with issuance of the patent (35 U. S. C. 154). Cf. Hoeltke v. Kemp Mfg. Co., 80 F. 2d 912, 923 (4th Cir. 1936).

250 Over the years the Patent Office has evolved numerous procedures to protect the parties to an interference. Principal among these is the practice of refusing access to filing dates until after "preliminary statements" giving dates of invention have been filed. See Patent Office Rules of Practice, rules 215–227. 251 Alexander Milburn Co. v. Davis-Bournonville Co., 270 U. S. 390 (1926). On the insufficiency of equiv. alent knowledge not in the form of a patent application see In re Schlittler and Uffer, 234 F. 2d 882 (C. C. P. A., June 21, 1956).