thermo-king-corporation-and-dawes-manufacturing-company ( 1961 )

292 F.2d 668

130 U.S.P.Q. 90

THERMO KING CORPORATION and Dawes Manufacturing Company,
Appellants-Appellees,
v.
WHITE'S TRUCKING SERVICE, INC. and Transicold Corporation,
Appellees-Appellants.
WHITE'S TRUCKING SERVICE, INC. and Transicold Corporation,
Appellees-Appellants,
v.
THERMO KING CORPORATION and Dawes Manufacturing Company,
Appellants-Appellees.

No. 18214.

United States Court of Appeals Fifth Circuit.

June 27, 1961.

Harold D. Field, Jr., Benedict Deinard, Minneapolis, Minn., Eugene C. Heiman, Miami, Fla., Leonard Street & Deinard, Minneapolis, Minn., Myers, Heiman & Kaplan, Miami, Fla., of counsel, for appellants.

Francis Browne, Washington, D.C., Bart L. Cohen, Miami, Fla., Schwarz & Cohen, Miami, Fla., William E. Schuyler, Jr., Andrew B. Beveridge, Mead, Browne, Schuyler & Beveridge, Washington, D.C., of counsel, for appellees.

Before RIVES, BROWN and WISDOM, Circuit Judges.

JOHN R. BROWN, Circuit Judge.

1

This case illustrates again that the shortest way through may be the longest way around. Through a combination of circumstances, none of which was finally within the control of either counsel nor caused by initial error on the part of the particular Judge trying the case, the Trial Court countenanced the complete disregard of a procedural statute. The loser contends that this error infected the whole proceeding so much so that its harmful consequences are clearly demonstrated. In any event, it is further urged, such a substantial doubt is cast upon the reliability of the Court's findings that a retrial free from the influences of this error is imperatively required. We agree and reverse.

2

The suit was by the patentee, Thermo King, against Transicold1 for infringement of three patents relating to mechanical refrigeration of automotive trucks.2

3

The Trial Judge, in his own words, 'reluctantly' held the patents valid but then found no infringement. The Patentee was denied all relief and the alleged Infringer was granted a sweeping injunction against further suits anywhere, anytime at any place against Transicold or any of its dealers or customers.

4

Since we do not deal with the substantive merits of the controversy, it is enough to describe the patents very broadly. A basic problem in refrigeration of trucks is to provide some means by which the operation of a freezer unit may be effectively regulated to maintain desired temperature levels. An automatic mechanical system is needed because of the practical inability of the truck driver, whose hands are already full, to care for this manually. In this area, the most acute single problem is defrosting the evaporator coils. Heat has to be somehow applied to the coils. And since the automatic systems use a gas refrigerant, whatever plan is employed has to assure that condensed liquids do not get back to the inlet side of the gas compressor.

5

These patents, and especially No. 2, are hailed by the Patentee as revolutionary discoveries which found the answers to the problems which so long made mechanical automotive refrigeration impracticable. No. 1 is a temperature control device that regulates the gasoline compressor engine at idle or full speed. No. 2 is a means of introducing so-called super-heat gas out of the stream of compressed gas into the evaporator coils to melt the accumulated frost on the exterior of the coils. This includes, of course, devices actuated by pressure or temperature differential controls to begin and end the defrost operation. No 3 discloses primarily an electric control circuit for use in a No. 2 super-heat system.

6

The pivotal factor is the notice provision of 282 of the Patent Act, 35 U.S.C.A. 282. This section first provides that the patent, when issued, is presumptively valid. It then permits, if pleaded, the defenses of non-infringement, invalidity for failure to comply with any of the specified requirements of the Act or any other defense specified in the Act. Of decisive importance here is the last portion which provides that in actions involving validity or infringement, the party asserting invalidity or non-infringement must give 30-day written notice of the prior art.3

7

The defendant-infringer, both in answer and counter-claim, denied validity and infringement. In the most general terms, it alleged that the patentee, Jones, 'was not * * * the inventor * * *'; that the claimed inventions 'were known or used by others in this country, or patented or described in a printed publication * * *' prior to Jones; that they 'were patented or described in printed publications * * * and/or were in public use or on sale * * * more than one year prior to * * * application * * * by * * * Jones * * *'; they 'were described in patents granted * * * others * * * before * * * Jones'; prior to Jones the inventions claimed were 'made in this country by others who have not abandoned, suppressed or concealed them;' the differences between prior art and invention 'would have been obvious to a person having ordinary skill in the * * * arts * * *' of 'the refrigeration and/or electrical arts'; the claims are inadequate under 35 U.S.C.A. 112 for want of particularity.

8

Not a single prior patent was cited in the infringer's pleadings and down to the commencement of trial no written notice of any kind was given of prior patents, prior publications, prior use, or prior art.

9

In many ways this case is the unintended victim of court congestion typical of many metropolitan areas such as Miami. It demonstrates also that the use of designated visiting judges not only does not offer a panacea, but brings on some special problems of its own. Experience has indicated that for efficient utilization of visiting judges, the resident presiding judge must make up the calendar in advance to supply a docket of cases fixed and ready for trial. Of course, the visiting judge still has the responsibility of taking action as subsequent developments require, but there is an understandable reluctance on his part to continue or postpone or abjourn a case thus fixed. Since visiting judges normally are designated for a limited period of time, to do so frequently means that the case must then fall back on the overcongested resident judge or upon some later visiting judge. Of course, one practical safeguard is an adequate and thorough pretrial to determine whether the case is really ready. These problems in court administration have much to do with the action in this case.

10

This case was not really very old.

11

The complaint was filed August 5, 1958, amended August 14. The defendants answered and counter-claimed on November 4, 1958. On December 22, 1958, the patentee filed its responsive pleading. On January 15, 1959, the Clerk sent written notice fixing the trial for February 15, 1959. Counsel did not receive this until January 17. Neither party was ready as the joint motion for continuance of January 28, 1959, attested. This was overruled by the presiding resident judge February 5.4 On the date (January 17, 1959) counsel received notice of the setting for February 15, there were, of course, less than 30 days remaining. The defendant could not have complied with 282. But no effort was made to give as much notice as the remaining time made possible. There were other pretrial maneuvers before the Court but in none was the possibility suggested that defendant had specific prior patents and prior art to offer on the trial.

12

The trial commenced February 18, 1959, before the visiting judge.5 In the opening moments of the trial the defendants for the first time tendered a written notice under 282.

13

That written notice listed 8 patents, 6 pieces of literature from the carrier corporation, 1 from another refrigerator manufacturer, and the names (and cities but no addresses) of two persons under the broad heading of 'Prior Invention Knowledge or Use.' The patentee objected vigorously to this belated tender. Its counsel pointed out the difficulties, if not impossibilities, of obtaining file wrapper histories and other data from the patent office and stated emphatically that it 'is utterly impossible' and 'we could not be afforded a fair trial.'

14

Except to recite the short notice of trial date, the unsuccessful mutual efforts to postpone the trial, and the relative newness of the case, the defendant could offer no reasons why notice had been so long delayed. Defendants called the Court's attention to Fairchild v. Poe, 5 Cir., 1958, 259 F.2d 329, 332-333, urged before us as well 'in support of the court's invoking its discretion and waiving the requirement of the rule.' Defendants offered also that 'subsequent to trial we will be happy to allow any amount of time that the plaintiff may wish.' The visiting trial judge after hearing this extended argument rejected all of these suggestions by the defendants and sustained the plaintiff's objection without qualification of any kind. The judge's reasons went to the very heart of the statute. 'It wouldn't be fair to require the plaintiff to meet this.'6

15

By this ruling the Patentee was informed that the trial would proceed without the necessity of meeting any prior art evidence of the type which the notice statute 282 and the case law construing it required a defendant to specify in writing. The defendants understood the full import of it, for they moved immediately for a continuance. The Court denied this out of hand.

16

But the trial, then just in its opening phases, was to turn out to be something quite different. For by the time the final going was rung in this 9-day trial, the judge admitted over vigorous objection of the Patentee, a total of 34 patents7 not a single one of which had ever been cited by the defendants before the trial began. In addition the Court allowed prior use testimony of two of the defendant's witnesses Swinburne and Kirkpatrick and supporting prior printed publication exhibits specified in the rejected belated written notice.

17

The admission of these specific patents in evidence cannot be passed off as cumulative or innocuous. The trial judge's lengthy formal opinion lists 20 patents as being relied on by defendants to establish the prior art. These 20 so listed include all four of Class D and six out of the seven of Class B, note 7, supra. Of the 20, he discussed at some length ten patents. These ten included five out of the seven of Class B and one out of the four of Class D. That means that the judge thought significant five patents out of the same Patent Office classification, but of which neither Thermo King nor its counsel had knowledge and at least one patent neither in the same class nor within Thermo King's actual knowledge. Also treated as significant were two of the patents in group 2 and two in group C as patents cited against patent No. 2 or against other Thermo King patents not in suit. How much the whole trial changed from its opening character is graphically portrayed by comparing this list of 34 patents with the belated notice tendered at the start of the trial. That list specified eight patents six of which were in the list of 34 admitted in evidence.

18

It is difficult to understand how the judge was ever led to admit those patents in groups B and D. The record never reflects any specific reason. Presumably the bars were let down altogether once the judge, perhaps unwittingly changing the rules in the middle of the game, Laird v. Air Carrier Engine Service, Inc., 5 Cir., 1959, 263 F.2d 948, reached the conclusion that those in groups 1, 2, 3, A and perhaps C were admissible despite 282. We think this pinpoints the source of the judge's error. Apparently he reasoned that 282 was intended merely to alert a patentee to the existence of prior art of which he would not have knowledge. Consequently, the argument continues, since a patentee knows about the prior patent cited by the patent office against the patent (or its predecessor applications), there is no need to give the statutory notice.

19

But this statutory notice has a more fundamental purpose than that, important though that be. In a day when we unconsciously think that pretrial discovery and preparation is a latter day contribution to jurisprudence, this statute is indeed interesting. Section 282 is an extension of former 69, formerly R.S. 4920, and traces its origin actually back to the Patent Act of 1790. Liquidometer Corporation v. Capital Airlines, Inc., D.C.D.Del.1959, 24 F.R.D. 319, 321. About one predecessor statute, 15 of the Patent Act of 1836, Justice Story had this to say: 'The object of this most salutary provision is to prevent patentees being surprised at the trial of the cause by evidence of a nature which they could not be presumed to know, or be prepared to meet, and thereby to subject them either to most expensive delay, or to a loss of their cause * * *.' Philadelphia & Trenton R. Co. v. Stimpson, 1840, 14 Pet. 448, 459, 39 U.S. 448, 459, 10 L.Ed. 535.

20

Intervening events have perhaps been a transfusion of restore full vigor to it. After the adoption of the Federal Rules of Civil Procedure in 1938, many courts held that the discovery rules superseded 69. Liquidometer Corporation v. Capital Airlines, Inc., D.C., 24 F.R.D. 319, 321 and note 4. But by the Patent Act of 1952, Congress deliberately intended to revive the statute as an effective regulation of patent trial procedure.8 The result is that the statute and the Federal Rules on discovery now coexist. The one does not exclude or supersede the other. The statute, subject always, of course, to the discretion reserved to the court to allow matters not specified in a notice to be received 'on such terms as the court requires' operates as a minimum. Liquidometer Corporation v. Capital Airlines, Inc., supra, 24 F.R.D. 319, at page 323.

21

In its deliberate re-enactment, the additions to the statute reflect an increased, not lessened, importance in its application. As re-enacted it covers both the defenses of invalidity and non-infringement. It was also expanded to cover not only directly anticipatory prior art references, but also prior art references to show the state of the art. And, of great importance, it added the last sentence, see note 3, supra, investing the trial court with discretion to admit non-noticed matters on such terms as the court requires. The objective of this statute in its expanded version has been summed up in C. S. Johnson Co. v. Stromberg, 9 Cir., 1957, 242 F.2d 793, 797. 'It is obvious that the purpose of the thirty day notice rule is to prevent surprise at the trial; to enable both sides to better understand adversarys' case; and to cut down on 'expensive delays,' or to prevent 'a loss of a cause' (Philadelphia & T.R. Co. v. Stimpson, 1840, 14 Pet. 448, at page 459, 39 U.S. 448, at page 459, 10 L.Ed. 535). The very purpose of the exception, as cast by Congress, was to permit and aid the ascertainment of the truth 'on such terms as the court requires.' * * * The notice provision in the statute is salutary. It should be encouraged * * *.' 3 Walker, Patents 660 at 1966, 1968 (Deller ed. 1937), Vol. 4 887D at 2768, 2771 and 1960 supp. to Vol. 4 at 58; see also Gere v. Canal Boiler Works, D.C.Wash., 33 F.Supp. 558.

22

This statute, just as interrogatories, requests for admissions, and the like under the Rules, is intended to do more than alert an adversary to the existence of evidence. It enables the plaintiff to know what sort of defense is going to be asserted. In a very real sense it determines what is to be tried. It has like value to the court either in trial or in the management of the serious administrative problems in determining what cases are ready for trial and capable of being submitted within the available time. The nature of patent litigation makes this of unusual importance. Judges trained in the law are confronted with baffling devices applying esoteric principles of all branches of higher mathematics and science. Long before the current interest in demonstrative evidence as it is sometimes used (or misused) in damage suits, the patent bar, with its generally high state of professional competence and extraordinary thoroughness in preparation, has found helpful ways to make the judges' task manageable and intelligible. Large multi-colored elaborate charts, diagrams, models and other exhibits, oftentimes costly and complicated, are prepared. They are used both on examination and serve with unusual effectiveness in cross-examination of experts and others skilled in the particular field. To the problem of understanding the substantive nature of the patented device or method may be added the further, if not more basic, one of communication: just what do these words-- often a long, prolix combination with a liberal sprinkling of adverbs and almost no punctuation9 -- of this claim mean? It is seldom safe to be a literalist. Bryan v. Sid W. Richardson, Inc., 5 Cir., 1958, 254 F.2d 191, 192. The meaning generally calls for a search into the practical considerations of the particular art. Inglett & Co., Inc. v. Everglades Fertilizer Co., 5 Cir., 1958, 255 F.2d 342, 347-48. This is a necessary consequence of the handy apothegm that a patentee is his own grammarian and lexicographer. And all of this naturally leads to the use of experts who, matching element by element,10 either as between the accused device and the patent, or as between the patent in suit and those asserted to anticipate, undertake to translate the problem into understandable, tangible, specific terms for the judge.

23

This whole search for truth in a specialized field where matters seldom turn on credibility in its cruder sense-- who is telling the unvarnished truth?-- is impeded unless the adversaries are fully prepared. Preparation, quite obviously, requires knowledge of what principal, specific, factual issues are to be examined.

24

Hence it is unrealistic to suggest as does the defendant that the Patentee could have overcome the impact of the court's disregard of 282 by asking for a continuance at the trial's end in order to meet or refute the prior art evidence admitted. There is, first, no indication that it would have been granted. The court had already denied two such motions, the last one in the context of this very problem. But in any event, a trial is something more than a truncated parade of one group of witnesses who swear 'yea' and another who respond 'nay.' The time to demonstrate effectively that the factual thesis advanced by a witness is not worthy of acceptance is while he is on the stand. But cross-examination in a technical area frequently requires the availability of material or at least effective demonstrative aids. Consequently, it is one thing to assert as do the defendants that counsel for the Patentee knew of a patent or a series of patents, or for that matter perhaps all of the patents in this patent office classification.11 It is quite a different matter to contend that by reason of such imputed knowledge such counsel, on the opening of the trial, had to contemplate that every one of such patents would be the subject of close examination and adversary contention.12

25

Nothing we have said is to be understood as holding that whenever 282 is breached the judgment must be reversed. We have previously stated, and now reaffirm, the basic proposition that much discretion is committed to the trial judge. Fairchild v. Poe, 5 Cir., 1958, 259 F.2d 329, 333; Edward Valves, Inc. v. Cameron Iron Works, Inc., 5 Cir., 1961, 286 F.2d 933, 949. From like considerations the statute itself allows evidence to be admitted 'on such terms as the Court requires.' What we deal with here is the same one faced where procedural rules are not followed. On the one side is the command of the statute, reflected in the Rules themselves that cases are not to be reversed if such error is harmless, F.R.Civ.P. 61, 28 U.S.C.A.; 28 U.S.C.A. 2111.

26

But on the other side is the positive principle expressed in the rule under scrutiny reflecting deliberate judgment by the Judiciary or Congress, or both, of those considerations which experience teaches to be of relevant significance in the administration of justice. Those aims-- those objectives-- become then a part of the inquiry whether transgression of the rule has been hurtful. As in any other area of court administration depending so much on the wise and sound discretion of the judge as he presides over the process leading finally to adjudication, the element of judgment cannot be supplanted by some ready mechanical formula. Hence, it would be a mistake to attempt to simplify it by casting the problem in terms either of who has the burden of demonstrating injury or proposing terms, or the necessity of demonstrating affirmatively that harm has resulted. As a rule 282 operates directly on the court. It is the court which is to impose terms as a condition to admissibility. The judge may not, therefore, passively wait on counsel. At the same time, a wise judge depends greatly on counsel and their suggestions on terms is deserving of a careful study. Likewise, on the element of harmfulness, the situation rarely permits an objective demonstration. Where the operation of the rule bears directly, as it does in 282, on the simultaneous fact impressions being made by evidence on the judge's mind, there is hardly any means of proving that the conclusions would have been otherwise. Indeed, erroneous exclusion of evidence does not demand such showing as a condition of reversal. It is sufficient if in context the evidence excluded has such persuasiveness that it might have led to a different result. In this aspect a factor of great importance is the insulation which surrounds fact findings by the judge as they come to us with the buckler and shield of F.R.Civ.P. 52(a). That so much is committed to the judge as he makes fact findings which may well dispose of serious controversy over complex and baffling patents makes it essential that in the fact finding process the judge, in a fundamental way, has applied those rules which lead to the full development of the truth.

27

Violation of procedural rules does not often compel reversal. But where we are left with the certain conviction, or more often an abiding impression, that a party was not afforded a fair and adequate opportunity to present his cause, we do not hesitate to reverse. See, e.g., Laird v. Air Carrier Engine Service, Inc., 5 Cir., 1959, 263 F.2d 948; Mitchell v. Johnson, 5 Cir., 1960, 274 F.2d 394; cf. Mitchell v. E-Z Way Towers, Inc., 5 Cir., 1959, 269 F.2d 126.

28

We have purposely refrained from discussing the merits lest something said or unsaid might influence decision on the merits as the case undergoes the new trial which we order. This litigation presents matters of great importance with substantial consequences in terms of money to these parties. Moreover it is a matter of great concern to the public interest. Consequently, we take pains to make plain that nothing said or unsaid, nothing included or omitted, nothing implied or expressed is to be regarded as a determination or even an intimation, one way or the other, on how this Court thinks this case or any of its numerous issues should be determined. The slate is literally wiped clean. The trial is about to commence. When the new trial is had and judgment reached, but not before, will it be for us to say whether the case ought to have been decided in a particular way.

29

Reversed and remanded for a new trial.

1

The patentee, Thermo King Corporation, was joined by Dawes Manufacturing Company, a partial licensee or assignee. The infringer, Transicold Corporation, was joined with a local Miami installation company, White's Trucking Service, Inc

2

In a commendable effort to overcome the sometime unavoidable obscurity of patentese, Bryan v. Sid W. Richardson, Inc., 5 Cir., 1958, 254 F.2d 191, 192, the record and briefs refer to the patents as Nos. 1, 2 and 3. They are:

Patent No. 1: Jones 2,477,377, 'Means for Thermostatically Operating Gas Engines,' issued July 26, 1949. Claims 1 to 4 inclusive.

Patent No. 2: Jones 2,666,298, 'Method and Means of Defrosting a Cold Diffuser,' issued January 19, 1954. Claim 6.

Patent No. 3: Jones 2,509,099, 'System for Controlling the Operation of Refrigeration Unites,' issued May 23, 1950. Claims 1 to 5 inclusive, and 13.

3

35 U.S.C.A. 282:

'* * * In actions involving the validity or infringement of a patent the party asserting invalidity or noninfringement shall give notice in the pleadings or otherwise in writing to the adverse party at least thirty days before the trial, of the country, number, date, and name of the patentee of any patent, the title, date, and page numbers of any publication to be relied upon as anticipation of the patent in suit or, except in actions in the United States Court of Claims, as showing the state of the art, and the name and address of any person who may be relied upon as the prior inventor or as having prior knowledge of or as having previously used or offered for sale the invention of the patent in suit. In the absence of such notice proof of the said matters may not be made at the trial except on such terms as the court requires.'

4

We treat this merely as a relevant circumstance and neither the parties nor this Court now state that this action was itself an error

5

The trial ran February 18-20, February 24-27 and March 2-3 with the case being taken under advisement on March 3, 1959

6

This was preceded by the following:

'The Court: The objection to the filing of this notice will be sustained.

'Counsel for defendants: May I say something?

'The Court: No. The Court has ruled. You can't come in here and-- you have submitted a list here that would require 30 days' notice for him to get ready for trial. This case is set for trial. Now, you have had since last October to file this very notice that you have given.

'I am sorry, gentlemen. I don't like to discommode you, but when you are set for trial you should be ready for trial and this notice is entirely too late. It wouldn't be fair to require the plaintiff to meet this.'

7

These have been grouped for convenience in the Patentee's brief roughly as follows:

No. of
Patents
Admitted  Group                    General Description
--------  -----  -------------------------------------------------------
4       1    Cited by Patent Office against Patent No. 1.
12       2    Cited by Patent Office against Patent No. 2.
4       3    Cited by Patent Office against Patent No. 3.
2       A    Cited by Patent Office against predecessor applications
of No. 2 patent.
3       C    Cited by Patent Office against other Thermo King
Patents (not in suit).
7       B    No knowledge by Thermo King--same Patent Office
class.
4       D    No knowledge by Thermo King--not same Patent Office
class.

8

See S.R. 1979, H.R. 1923, reprinted 1952 U.S.Code Congr. & Admin.News, Vol. 2, p. 2394. The Committee Reports adopt the appended revision notes as a part of the report, p. 2403. These notes state:

'The five defenses named in R.S. 4920 are omitted and replaced by a broader paragraph specifying defenses in general terms.

'The third paragraph, relating to notice of prior patents, publication and uses, is based on part of the last paragraph of R.S. 4920 which was superseded by the Federal Rules of Civil Procedure but which is reinstated with modifications.' page 2422.

The Commentary by Zinn, page 2507, at 2523, stated: 'The final paragraph of the section restates the provision requiring notice of the references to be relied upon and includes a reference to the state of the art. This is intended to avoid surprise at the trial.'

The Commentary on the New Patent Act, Federico, Examiner in Chief, U.S. Patent Office, Vol. 35 U.S.C.A. 1-110, p. 1, at 54-56, concludes: 'The last paragraph of section 282 relating to the giving of notice of various details relating to certain defenses is based on part of the last paragraph of old R.S. 4920, with modifications. The old provision was in fact superseded by the Federal Rules of Civil Procedure but, as modified, has been reinstated.'

9

As an illustration, see the 334-word claim in Thurber Corp. v. Fairchild Motor Corp., 5 Cir., 1959, 269 F.2d 841, 850, note 7. Such draftsmanship may be matched only in federal revenue statutes. See the breath-taking 593-word sentence in 26 U.S.C.A. 341(e)(1) concerning collapsible corporations. Much of this is simply unavoidable. Courts may bear a considerable responsibility in this area since hypercritical interpretations invite more, not less, prolixity as careful solicitors undertake to spell out a certainty that will dafy judge-made construction. The patent bar is alive to this problem of communication. See e.g., Lutz, Better Public Relations Through Better Patents, Patent Office Society Journal, Aug. 1960; Arthur M. Smith, The Art of Writing Readable Patents, 1958 Practicing Law Institute. And much is being done to simplify this problem of patentese by the mechanical construction and arrangement of the claims. There is a growing use of parenthetical lettered or numbered constituent elements, the typographical arrangement into indented, blocked subparagraphs, and the like. This recent recognition of what formerly might have been regarded by purists as rank heresy rests on the certainty that ultimately the claim will have to be read by strangers to its creation. Hence, why not make it manageable? See, for example, patents of these solicitors:

2,792,999  Charles F. Chisholm
2,958,454  Soans, Anderson,
Luedeka and Fitch
2,952,319  Jerry J. Dunlap
2,963,038  Jerry J. Dunlap
2,943,765  Arthur L. Wade
2,930,956  Arthur L. Wade
2,939,636  Brown, Critchlow,
Flick and Peckham
2,848,523
2,952,193  D. C. Snyder and
A. R. McCandy
2,960,706  W. Glenn Jones
2,956,864  Harris, Kiech,
Foster and Harris
2,978,637  Tom Arnold

Presumably the group of Primary Examiners had just such things in mind in their recent recommendations. 'Within the claims, the clauses thereof shall be separated, in order that the examiner may readily determine those individual factors which are essential to the claim.' XLIII JPOS 317 at 319.

10

We have pointed out how essential this is to our orderly underatanding of these complex devices. Bryan v. Garrett Oil Tools, 5 Cir., 1957, 245 F.2d 365, at page 374

11

These are the patents in group B, see note 7, supra, of which Thermo King had no knowledge but which were admitted because they were in the same Patent Office class

Imputation of such constructive knowledge has a dubious basis. For example, the briefs tell us without contradiction that in Patent Office class 62, Refrigeration, there are 75,000 patents divided into 531 subclasses.

This whole problem of classification of patents, as the issuance of nearly three million patents and an annual application for 75,000 more makes the physical task of search and analysis more difficult if not really impossible, is a matter of great concern to the Patent Office and the patent bar. Interestingly enough, concerted, successful experimental efforts toward a solution now serve as tangible hopefulness for those active in the proposed use of electronic data computors in legal research generally. The lead is being taken by the ABA Committee on Electronic Data Retrieval (EDR) whose activities are currently reported in MULL (Modern Uses of Logic in Law, a quarterly published at Yale University for EDR). See, generally, for bibliographies on machine assistance in legal research, see MULL 59S4; Lawlor, MULL 59D4b; and Hayden, MULL 60J4a and b; Biunno, History of Electronic Methods for Legal Research, MULL 60Sld; 'Applications of Electronic Data Processing Systems To Legal Research,' Proceedings of the Electronic Data Retrieval Committee of the Bar Activities Section of the American Bar Association, August 29, 1960, Washington, D.C., pp. 3-13, (Bureau of National Affairs): Dickerson, Electronic Law Searching, MULL 60Slb; Horty, Electronic Data Retrieval Techniques Surveyed, Virginia Law Weekly (February 16, 1961), pp. 2-3; Freed, Pushbutton Research: Automation In the Law Library, The Shingle (January 1961), p. 11; Chasalow, The UCLA National Law and Electronic Conference, reporting on The National Conference on Law and Electronics, University of California at Los Angeles, UCLA, October 1960, MULL 60Sle.

The spectacular accomplishment by the Patent Office always receives much attention. See, 'Experience with Electronic Searching of United States Patents' by Donald D. Andrews, United States Patent Office, Proceedings of the Electronic Data Retrieval Committee, supra; Newman, Information Retrieval Research in the U.S. Patent Office, MULL 60J45; 'Computers and Antitrust, Copyright and Patent Law' by Bartholomew Diggins, attorney Washington, Three-day Forum, Joint Committee Continuing Legal Education of the A.L.I. and the ABA, March 23-25, 1961.

The story of the success of the Patent Office can be pieced together from the following literature. For the skeptic it is good to read it in chronological order as this reflects the guarded hesitant proposed experiments, the success which invariably exceeded expectations, and the constant growth in application and complexity of the problems being programmed for machine handling together with adaptation to like improvements in the data computers themselves. Bailey, Lanham & Leibowitz, Mechanized Searching in the U.S. Patent Office, 35 J. Patent Office Soc'y, 566 (1953); Lanham & Leibowitz, Classification Searching and Mechanization in the U.S. Patent Office, 40 J. Patent Office Soc'y 86 (1958); Additional Steroid Punched Card File, 41 J. Patent Office Soc'y 295 (1959); The Annual Report of the Commissioner of Patents, Fiscal Year 1959, 42 J. Patent Office Soc'y 147 at 152-156 (1960). This Annual Report refers to staff reports of the Office of Research and Development of the Patent Office: No. 12, Storage and Retrieval of Contents of Technical Literature-- Nonchemical Information; No. 16, Analysis of Prepositionals for Interrelational Concepts-- Preliminary Study; No. 15, A Notation System for Transliterating Technical and Scientific Texts for Use in Data Processing Systems; No. 11, A Manual for Coding Steroids; No. 13, A System of Retrieval Compounds, Compositions, Processes, and Polymers; No. 14, Variable Scope Patent Searching by an Inverted File Technique.

Two other interesting articles in this area are, Lawlor, Analysis of Patent Claims by Mathematical Logic, Patent, Trademark, and Copyright Law Section, Amer. Bar Ass'n., 1961, and Freed, Prepare Now for Machine-Assisted Legal Research, 47 A.B.A.J. 764 (Aug. 1961).

12

A reading of this whole record bears out what all counsel acknowledged on argument before us. The judge's erosion of his opening refusal to allow reference to the patents not noticed, see note 6, supra, caused a frantic scramble with much long distance telephoning and burning of midnight oil to obtain copies of patents and related materials as more and more patents were gradually let in