Globe-Union, Inc. v. Chicago Telephone Supply Co. , 103 F.2d 722 ( 1939 )

103 F.2d 722 (1939)

GLOBE-UNION, Inc., et al.
v.
CHICAGO TELEPHONE SUPPLY CO. et al.

No. 6682.

Circuit Court of Appeals, Seventh Circuit.

April 4, 1939. Rehearing Denied May 19, 1939.

Frank H. Hubbard, of Milwaukee, Wis., for appellants Cutler-Hammer, Inc., George J. Meuer and William C. Stevens.

George I. Haight and M. K. Hobbs, both of Chicago, Ill., and John W. Michael and Gerrit D. Foster, both of Milwaukee, Wis., for other appellants.

George L. Wilkinson and Howard W. Hodgkins, both of Chicago, Ill., for appellees Chicago Telephone Co. and Newton Schellenger.

Before EVANS, MAJOR, and KERNER, Circuit Judges.

KERNER, Circuit Judge.

This appeal involves a contest among rival inventors^[1], who claim priority of invention in a certain rheostat-snap switch mechanism defined in a single count^[2] as follows:

"In combination a volume control comprising a circular resistance, a contact arm for varying said resistance, a rotatable actuator shaft on which said contact arm is mounted, a projecting pin eccentrically mounted on said shaft, a casing enclosing said volume control, an aperture in said casing, an auxiliary recessed casing closing said aperture, a snap-switch contained in said auxiliary casing, and an actuating arm for said snap-switch operable by said projection at one extreme of its cycle."

The three inventors involved filed applications^[3] in the United States Patent Office at about the same time: Schellenger's, serial No. 602006, on March 30, 1932; Stoekle's, serial No. 602855, on April 2, 1932, which was referred back to his prior application, serial No. 532138, on April 23, 1931; and Meuer's and Stevens', serial No. 537774, on May 16, 1931. An interference was declared and the case was heard by the Examiner of Interferences of the United States Patent Office, who awarded priority of invention to Schellenger. An appeal was taken to the Board of Appeals of the United States Patent Office, which affirmed the decision of the Examiner of Interferences, whereupon Stoekle and Globe-Union Inc., assignee of the Stoekle claim, brought an equity suit under R.S. § 4915, 35 U.S.C.A. § 63, to adjudicate the question of priority. This appeal was taken from the district court's decree^[4], which favored the Schellenger claim to priority of invention.

Disposition of this appeal depends on the answer to the following question: Which inventor was entitled to a patent on the combined volume control and switch-unit described in the co-pending applications before the Patent Office? To give response to this question, it is imperative that we in turn answer two factual issues: Was Schellenger, the first conceiver, the first to reduce his conception to practice? And, if Schellenger is not the first reducer, was he reasonably diligent in reducing his idea to practice? Since the district court found for Schellenger, it is necessary that Stoekle convince this court thoroughly that Schellenger was not entitled to a favorable answer to both questions. Morgan v. Daniels, 153 U.S. 120, 14 S.Ct. 772, 38 L.Ed. 657; Gold v. Gold, 7 Cir., 237 F. 84.

A suit in equity between interfering claims is an original action, in general independent of earlier Patent Office interferences, and the issues therein are tried de novo upon all competent evidence, old or new. Courson v. O'Connor, 7 Cir., 227 F. 890; Uihlein et al. v. General Electric Co., 7 Cir., 47 F.2d 997, 999. The old evidence, i. e., the testimony contained in the Patent Office record, follows:

Stoekle filed his application on April 23, 1931, and Schellenger filed his application on March 30, 1932. Stoekle fixed the date of conception as about September 16, 1929 and the date of reduction to practice on January 10, 1930, whereas Schellenger fixed conception on April 18, 1928 and reduction to practice as about May 13, 1928. The dates submitted by Stoekle are not disputed. On the other hand, dispute exists over the dates fixed by Schellenger. The entire record in the interference proceedings was by depositions and exhibits.

1928 work: Schellenger, electrical engineer for the Chicago Telephone Supply Company which manufactured radio supplies, testified that he conceived the built-in radio control device on April 18, 1928. For corroboration he pointed to his 1928 diary, the appropriate calendar entry of which revealed drawings of the improved art here in question. In addition, three supervisory employees of the company and Mrs. Schellenger testified that they saw the drawings of the 1928 diary in the spring of 1928, while sales executive Hill and President Best stated that Schellenger had disclosed his conception to them.

Schellenger then testified that on May 13, 1928 he reduced his concept to practice by making the combined rheostat switch and successfully installing it in his own personal battery-operated radio receiving set at home. This testimony is supported by the diary, the three supervisory employees (Kehres, Steely, McMeekan), and Mrs. Schellenger. Schellenger continued by stating that thereafter the control arrangement was removed to the factory to serve as a guide for his second sample and that on August 5 of 1929 the original device was given a life-test on 110-125 volt A. C. current. This life-test was described in the diary and the device was seen at the life-testing machine by two of the three supervisory employees, by one Smead (a former supervisory employee who had gone to the plant on that particular day to see about a job), and by one Tait (an employee who was told by Kehres that the unit he saw being tested was a combination rheostat and snap-switch).

On the other hand, neither the 1928 original device nor the 1928 second sample was produced in evidence, Schellenger explaining that they had been mislaid and that a diligent search had been made in vain. A duplicate of the 1928 original device, made several days before these depositions were taken, was offered in evidence, and described generally by the corroborating witnesses, Mrs. Schellenger and the three supervisory employees, as being similar to the original. Except the diary, no records were kept of the 1928 radio test and of the 1929 life-test. Moreover, these corroborating witnesses testified in October of 1934 from memory concerning events in 1928 and 1929.

1930 work: Schellenger testified that in 1930 little work was actually done on the control unit in question. He gave increased production tasks and bad health as reasons for this inactivity. This testimony was corroborated by other witnesses. Schellenger then stated that in the spring of 1930 he turned over the development of the rheostat switch unit to engineer Haselwood, who testified that he did not work much on the unit in 1930. Schellenger said that he gave verbal instructions and sketches to Haselwood, that he directed him to make drawings and samples, and that he gave greater thought to the whole matter during the latter part of 1930.

The record also showed that during 1928, 1929, 1930, and 1931 Schellenger and his company stressed the development, improvement and manufacture of the bracket or outrigger types of rheostat switch construction. Giving most of his time and activity to the production of the outrigger combinations (which were in demand by the trade), neglecting the built-in control unit (which admittedly was vastly superior), and saying that the new unit had been so developed by 1930 that a high school boy following instructions could have finished it, is also shown by the record in the interference proceedings. In addition, neither records nor drawings nor sketches were produced in evidence as representing what work was done in 1930.

1931 work: Schellenger testified that the final work on the rheostat switch device was started in January of 1931. Haselwood corroborated this testimony and stated that he had supervised the making and testing of samples in February, March, April, and May. Some of these samples were produced in evidence and the activity was supported by the 1931 diary. Abel and Veatch, employees who made the samples, also corroborated this testimony. Time cards of Abel and Veatch were produced which showed that in June of 1931 they had been working overtime, and late in June Haselwood obtained the approval of the samples by the Underwriters' Laboratories. The company began to manufacture the built-in rheostat switch in late summer of 1931.

The record revealed other evidence. Thus, Haselwood stated that in March Stoekle had advised Schellenger that Cutler-Hammer, manufacturer of bracket and barnacle (built-in) switches, was making exclusively for Stoekle's combination unit a "switch to go in the cover," and Schellenger testified that it was not until after he had completed his unit and sometime late in 1931 that he saw the Stoekle unit.

On the other hand, Phillips, sales executive for Cutler-Hammer, testified that sometime in April of 1931 Schellenger had expressed an interest in the barnacle or built-in switch which was then used in the Stoekle combination rheostat and snap-switch unit.^[5] Phillips introduced in evidence a letter from Schellenger, dated April 25, 1931, which acknowledged this interest in the following way:

"I have one of your new switches on my desk and must say that the same impresses me very favorably."

This letter was addressed to Greenman, Cutler-Hammer salesman, and Phillips did not have personal knowledge that the switch referred to in the Schellenger letter was the switch used by Stoekle.

Lastly, in explaining the delay in filing an application for a patent, Schellenger testified that the company was not "patent conscious" at the time and that therefore it was then their policy to delay until they had perfected the device commercially. Yet, during 1928, 1929, 1930, and 1931 at least 25 patents had been issued to the company with Schellenger appearing as inventor of most of the patented devices.

The Board of Appeals in the United States Patent Office found on the record related above that the 1928 work was a reduction to practice (not an abandoned experiment) and that Schellenger's subsequent conduct did not call for an application of the Mason v. Hepburn doctrine. See 13 App.D.C. 86. The diaries were referred to considerably throughout the decision; the Stoekle-Schellenger conversation in March of 1931 and the Schellenger letter of April 25, 1931 were not so relied upon. The new evidence, i. e., the additional testimony adduced in the district court, follows:

Stoekle introduced expert testimony which tended to discredit the diary entries relating to the rheostat switch in question here. Mrs. Keeler, Stoekle's handwriting expert, testified (1) that the entries relating to the invention by Schellenger were either the only or the last (bottom) entry upon the diary pages; (2) that there were many entries which were not made on the dates the diary pages bore; (3) that in many instances entries bearing the same date were made on different dates and with changing sequence; and (4) that Schellenger made the questioned entries, which cover a period of four years, all at one time.

Faxon, Schellenger's expert witness, agreed that the use of "pickups" and "transfers" of lead constituted a scientific basis for facts as to the relative order or sequence of the entries. That is, Faxon agrees that (2) and (3) (above) are proper conclusions. He disagrees, however, that (4) is a proper conclusion, as far as scientific study is concerned, while no one will deny the strength of conclusion (1).

That (4) is a proper deduction from the circumstances is not as clear as (1), (2), and (3). Our analysis of these entries, however, leads us to the conclusion that the entries are suspicious and therefore not reliable. The top entries on the pages of the three diaries fall easily in one group, in which the handwriting varies greatly. The bottom entries constitute another group and pertain invariably to the combination switch and rheostat: these entries are written with the same general appearance, showing little variation as to size and spacing of the words; they differ obviously from the top entries on the same pages, and, where offsets and pickups are possible, indicate that they were written out of order from the preceding and following top entries.

Stoekle then introduced Greenman and Phillips as witnesses testifying as to the Schellenger letter of April 25, 1931, Jonas and Miller as to Schellenger's activity in 1931, and McClain as to Schellenger's reduction to practice in 1928.

Greenman, Cutler-Hammer salesman, in the spring of 1931 had been trying to sell Schellenger a radio switch, and Schellenger expressed interest in the barnacle type of switch which Cutler-Hammer was making for Stoekle. Greenman, addressee of the Schellenger letter of April 25, 1931, the last paragraph of which acknowledged that Schellenger had the switch in his possession, confirmed Phillips' testimony that the switch mentioned in the letter referred to the barnacle or built-in type used by Stoekle in his unit. Jonas, who worked from 1928 to 1932 in the tool room and laboratory, testified that prior to April of 1931 he had never seen a built-in switch in the plant. He insisted that in April of 1931 Schellenger dissected the Cutler-Hammer barnacle type of switch in his presence, that he then observed draftsman Arisman making a drawing of the dissected switch, and that later he saw activity in the nature of preparation for the manufacture of the built-in switch. Miller, tool maker from 1927 to 1932, corroborated Jonas' testimony, and Haselwood on cross-examination admitted that he saw the Cutler-Hammer switch in the plant in April of 1931.

McClain, chief inspector from 1928 to 1935, testified that he operated Schellenger's radio in June of 1928 but that he did not notice a volume control device with a built-in switch; that the first time he saw it was in June or July of 1931 when Miller had it dissembled. McClain also stated that in the fall of 1934 Schellenger told him that he was in trouble over a patent and suggested that he should remember that he saw the device in the radio in 1928, in case it was necessary to use him as a witness. Schellenger denied this testimony and stated that he had refused McClain, when he volunteered to perjure himself in Schellenger's behalf. Haselwood and Steely corroborate Schellenger in this matter.

On the whole record, as related above, the district court decreed in favor of Schellenger, finding a reduction to practice in 1928 and excusable conduct subsequent to 1928. After indicating that it favored Schellenger, the court requested his counsel to propose findings of fact. With the exception of one, the court accepted the proposed findings as being true; the court rejected the proposed finding that the diaries in evidence were authentic and reliable.

In our analysis of the whole record, we can not escape the strength and the compelling influence of the additional evidence that was adduced in the district court. This evidence undermines the foundation of the proof that Schellenger had reduced his conception to practice in 1928, and consequently throws a new light upon the whole case.

We appreciate that the principle of Morgan v. Daniels, 153 U.S. 120, 14 S.Ct. 772, 38 L.Ed. 657, controls in equity suits brought under R.S. § 4915, 35 U.S.C.A. § 63. In Uihlein v. General Electric Co., 7 Cir., 47 F.2d 997, on page 1004, this court said that the statute, and the decisions thereunder, had placed upon the plaintiff a "heavy burden, which could only be overcome ``by testimony which in character and amount carried thorough conviction' and which left the court free from reasonable doubt."

The "burden", as used above, refers to great weight which attaches, and should attach, to the decisions of the Patent Office. The final determination of another department of the Government is not lightly to be set aside by the courts. Thus, in Imperial Brass Mfg. Co. v. Hackney, 7 Cir., 75 F.2d 689, 691, we approved of the wholesome instruction of the Supreme Court in Morgan v. Daniels in this fashion:

"But where the contested issue in the two proceedings (the interference and the infringement suit) is the same, and the disputed issue of fact is identical, why shouldn't the determination of the contest in the Patent Office assume well nigh dominating importance in the determination of the infringement suit? Grant that the ruling of the examiner is not res adjudicata — that the finding in the interference contest may be shown to be erroneous — yet the rule applies, namely that where the evidence is practically the same and the issues are identical, and the determination of the issues turns upon the question of veracity, the second trier of fact must of necessity give much weight to the finding of the first trier of fact."

In the mine-run of cases that come up through the Patent Office the record of the trial court is a mere repetition of the record of the Patent Office. In such situations, i. e., where the new evidence is not important or at most cumulative in nature, it is only reasonable that the finding of the Patent Office should assume "well nigh dominating importance" in the trial court. An analysis of the facts in Morgan v. Daniels, Uihlein v. General Electric Co., and Imperial Brass Mfg. Co. v. Hackney reveals clearly that the two records, i. e., that of the Patent Office and that of the trial court, were practically the same in quality and quantity.

The instant case is different. The two records involved are not the same. The additional evidence found in the trial court's record is not merely cumulative in nature. On the contrary, it carries with it thorough conviction, and strikes hard at the core of Schellenger's evidence. This becomes very important, when we realize that the determination of the issues in these cases is ordinarily not dependent upon the question of credibility of the witnesses so much as on the compelling force of the stories told by the depositions, which make up the bulk of the record.

Counsel for Schellenger contends that the additional evidence taken in the district court should have been disregarded on the ground that it was available to the Patent Office. Barrett Co. v. Koppers Co., 3 Cir., 22 F.2d 395; Greene v. Beidler, 2 Cir., 58 F.2d 207; Perkins v. Lawrence Sperry Aircraft Co., D.C., 57 F.2d 719; O'Donnell v. United Shoe Mach. Corp., D.C., 2 F. Supp. 178. Since the last three cases cited refer specifically to the principle announced in Barrett Co. v. Koppers Co., it is not essential to pay further attention to them. The court in the Barrett case said [22 F.2d 397]:

"In a contest between two claimants to the same invention on an issue of priority each is expected to produce all the testimony he has on that issue so that the Patent Office tribunals and the courts may make right decisions. If for some reason of his own a party withhold evidence which is available to him and which he can produce at will but does not produce, then he must be regarded as having abandoned that evidence in its bearing on the issue under trial. When that issue is decided it is somewhat in the nature of res judicata as to the evidence withheld."

We do not deny the logic and expediency that underlines this statement of the court in the Barrett case. We do deny the applicability of the principle in the instant case.

In the Barrett case, the additional evidence was at all times "wholly within [the] possession and control" of plaintiff, but he had "withheld" the evidence from the Patent Office. Under such circumstances, it is reasonable that the withholding person be estopped to present that evidence later. Such a principle of estoppel works properly in many instances, e. g., when evidence has been deliberately withheld or secreted, and when a story is completely changed on coming to court. We do not dispute the soundness of the proposition that all pertinent evidence, actually available, should be submitted in the first instance. To permit partial presentation before the Patent Office is to sanction the destruction of administrative justice.

But we are satisfied that the estoppel principle has its limitations. It should not be used to penalize an innocent party. Nor should it exclude the presentation of evidence, which previously had not been procurable or which had become known after the interference proceedings. Coming now to the instant case, we find that Stoekle did not intentionally withhold the evidence which he later produced in court. The mere fact that he sought to discredit the diary entries in the interference proceedings militates against the claims of secretion or intentional withholding. It seems undisputable that evidence may be available and existent at a given time, and yet not be within the knowledge or "possession and control" of the person desiring to make use of it. We are not prepared to say that Stoekle did not exercise due diligence in procuring the evidence sooner than he did.

Counsel goes further and insists that the court in the Barrett case did not restrict the principle announced to situations where the evidence was deliberately withheld or otherwise procurable by the use of due diligence. We disagree with counsel. Every opinion must be read with reference to the facts upon which it is based. The Barrett case stands on facts clearly showing intentional withholding of evidence. In fact, even if counsel were correct, we would not go that far. To hold with counsel's contention is tantamount to judicial abrogation of R.S. § 4915, 35 U.S.C.A. § 63. Such a contention would make it impossible for a plaintiff to meet the test laid down by the doctrine of Morgan v. Daniels, i. e., that the interference-judgment can be overcome "by testimony which in character and amount carries thorough conviction." [153 U.S. 120, 14 S.Ct. 773.] We must always be conscious of the distinction between a de novo trial, where the case is tried a second time and the record is made up in the district court, and a court's review of findings of an administrative body, where the record is solely that of the administrative body. The instant case falls in the former class.

In the case at hand we have a situation in which the junior applicant, Schellenger, claims he is prior conceiver and prior reducer to practice. The law is plain that in such cases the junior applicant carries the burden of proving clearly that he antedates the senior applicant. Judge L. Hand, in United Shoe Mach. Corp. v. Brooklyn Wood Heel Corp., 2 Cir., 77 F.2d 263, 264, expressed the applicable law in this way:

"When an inventor's date is to be carried back beyond his application, courts regard the effort with great jealousy, and must be persuaded with a certainty which is seldom demanded elsewhere; quite as absolute as in a criminal case, in practice perhaps even more so."

Schellenger attempted to meet this burden by evidence that he conceived and reduced his idea to practice in 1928. His proof was based on three diaries and corroborating oral testimony. The Board of Appeals in the Patent Office held that Schellenger reduced his idea to practice in 1928. Although we are not positive that the Board relied solely on the diary evidence, it is clear that the Board accepted and relied to great extent on the pertinent diary entries, for its decision is replete with references to the diaries. The Board, in denying a petition to reconsider its decision, stated that "Petitioner also attempts to discredit the entries in Schellenger's diary but this matter was considered before rendering our former decision and there is no new argument advanced in this connection."

Then, in the district court, Stoekle produced additional evidence which cast a heavy shadow of suspicion over the pertinent diary entries. Our thorough study of the diaries convinces us that the pertinent diary entries are not reliable, and that Schellenger must carry on without their use. If Schellenger's oral testimony is insufficient to meet the burden described by Judge L. Hand in the United Shoe Machinery case, supra, it follows that Stoekle's additional evidence, going to the heart of Schellenger's proof, has overcome the interference-judgment by "testimony which in character and amount carries thorough conviction." Even though our condemnation of the diaries might very well shake the whole evidence structure, our conviction teaches us that to lend reliance to the diary evidence after Stoekle's attack is to do violence to our common sense and reason.

The question paramount in importance is whether Schellenger, admittedly the first conceiver, reduced his conception to practice in 1928. The situation comes down to this: The conceiver, three supervisory employees of the company, and his wife testify that he completed the 1928 radio unit three weeks after date of conception, installed it in his radio on May 13, and operated it successfully for the major portion of the summer.

Save for the diaries, the record is devoid of contemporaneous physical or documentary evidence of activity which, according to the oral proof, occurred in 1928, 1929, and 1930. This in itself is extremely incredible, when considered in light of the complete story concerning the two rival inventors. It is difficult to believe that Schellenger, who methodically kept a diary for years and who manifestly was capable of protecting any of his inventions, did not keep one independent record or document during these three years. The truth is that he neither kept the 1928 radio device nor preserved records of the radio test and life-test. He mentioned that he started a second sample in 1928, but he was unable to produce this also. This lack of physical and documentary proof contrasts strangely with the abundance of such proof concerning his 1931 activity. We are not satisfied that Schellenger embodied his idea in a practically operative instrument in 1928.

As to the oral evidence, it is important to remember that the Supreme Court has spoken several times on the inherent dangers of unsupported oral proof in this class of cases. In the Barbed Wire Patent, 143 U.S. 275, 284, 12 S.Ct. 443, 447, 450, 36 L.Ed. 154, the court made this pertinent comment:

"In view of the unsatisfactory character of such testimony [oral testimony], arising from the forgetfulness of witnesses, their liability to mistakes, their proneness to recollect things as the party calling them would have them recollect them, aside from the temptation to actual perjury, courts have not only imposed upon defendants the burden of proving such devices, but have required that the proof shall be clear, satisfactory and beyond a reasonable doubt. * * * The very fact, which courts as well as the public have not failed to recognize, that almost every important patent, from the cotton gin of Whitney to the one under consideration, has been attacked by the testimony of witnesses who imagined they had made similar discoveries long before the patentee had claimed to have invented his device, has tended to throw a certain amount of discredit upon all that class of evidence, and to demand that it be subjected to the closest scrutiny."

In this case, which is already shrouded in shadows of suspicion, there is reason to apply the "closest scrutiny." We are satisfied that Schellenger and his four main corroborating witnesses do not meet the scrutiny test.

The record is plain that the three supervisory employees and Mrs. Schellenger were not well-advised concerning the exact nature and structure of the 1928 radio device, which they testified they saw in the Schellenger radio in the summer of 1928. Of course, this is to be expected; six years had elapsed since then to dim their recollection. This oral testimony, general at most, illustrates the uncertainty which invariably attaches to the testimony of an ordinary witness, who speaks on a subject-matter that falls within the province of inventive genius. We are, therefore, not ready to say that after such a long interval of time this evidence, unsupported as it is by physical or documentary proof, carries convincing weight.

We are also unable to give much credence to Schellenger's oral testimony, for it can not be dissociated from his many references to, and his refreshments of memory from, the diaries. In this case, there is also a greater temptation to color the facts in his favor. For instance, in the interference proceedings, Schellenger firmly denied that he had seen his rival's radio unit before the fall of 1931. In the district court, the testimony of Greenman and Phillips, reinforced by Haselwood's admission on cross-examination, conclusively refuted Schellenger in this regard. We would be less certain in our conclusion, if Schellenger's subsequent conduct was corroborative of, instead of inconsistent with, his oral testimony that he had successfully operated the Schellenger unit in 1928.

At this time it is proper to dispose of counsel's contention that the record contains documentary proof of the activity in 1928. Counsel argues that "The tests were recorded in the diary and the making of the tests was fully corroborated by the witnesses, who fixed the dates independently of the diaries." The strength of this argument does not appeal to us. To us this seems to be trying to do indirectly what can not be done directly. Common sense teaches us a different lesson: the oral testimony does not purge the diaries so much as the diaries pollute the oral testimony.

However, we do not go so far as to reject the entire oral testimony given, but, giving it all the weight to which it can reasonably be entitled, it shows no more than uncertain experimentation. The unsatisfactory oral proof, the unreliable diary evidence, and the lack of test records and physical proofs of the "successful" 1928 device convinces us that Schellenger did not actually embody his idea in a successfully operative device in 1928. His subsequent inactivity regarding the 1928 device, accompanied by his activity concerning the improvement of the outrigger type of construction, which was the forerunner of the barnacle type of construction here in question, is not consistent with reduction to practice in 1928.

It is inconceivable that Schellenger, extremely capable of taking advantage of any invention he might make and admittedly endeavoring to evolve a more successful radio rheostat and switch, would have acted as he did in 1929 and 1930, had the 1928 radio test been measurably successful. Let us assume that Schellenger succesfully operated the 1928 device in his own radio during the summer; that a "good bright high school boy", following his instructions, could have perfected it easily thereafter; that it was perfected in a short time in 1931; and that it was superior to and outmoded the outrigger device. If this were true, it taxes the power of belief that in 1929 and 1930 he directly turned his major attention to the improvement of the inferior outrigger unit.

The only logical conclusion is that Schellenger proved uncertain experimentation in 1928, which is not sufficient. The law is clear in such instances. The prior conceiver, who slumbers on his rights, must have very clear and convincing evidence that the test fully complied with the requirements as to reduction to practice, and that if he fails to sustain that burden the test will be regarded as mere abandoned or uncertain experimentation, and priority awarded the party through whose efforts the public received the invention.

We come now to the question of diligence. It is an elementary proposition that a first conceiver must use reasonable diligence in reducing his idea to practice, in order to entitle him to a patent as against a subsequent conceiver who has first produced the concrete art. The law, however, has not established a general standard by which diligence can be measured. All that can be said on the subject is that the conceiver's conduct between the conception and the reduction must be reasonable under all the circumstances of the particular case in question. In the instant case, since we have found that he did not reduce his idea to practice in 1928, Schellenger is the prior conceiver, conceiving in 1928, and the last reducer, reducing his conception to practice in 1931. We are satisfied that Schellenger's conduct in the interval between 1928 and 1931 did not constitute reasonable diligence in effecting reduction of his conception to practice.

Giving Schellenger's testimony the most reasonable weight possible, it shows no more than this: Schellenger had conceived the idea in 1928, and had experimented for sometime during the summer of 1928. Although he had not forgotten about the idea, it rested in uncertain experimentation, and there was great uncertainty as to whether any more activity should be devoted to it; after almost three years of inactivity, he is suddenly spurred into great activity by Stoekle and he reduces the conception to practice. No other reasonable theory fits the facts in this record.

We are convinced that from the fall of 1928 to the spring of 1931 Schellenger did not do all that he could have done, nor do we believe that he was hindered in reducing his idea to practice by circumstances beyond his control. There is testimony in the record as to his general bad health, but there is also testimony from the same source that his production tasks and activities in general had increased. In 1929 and 1930 his emphasis and energy were directed to the development and improvement of the bracket type of switch in their outrigger unit, which represented the first step in the combination of the volume control and switch; although he may not have forgotten about his conception of the barnacle type of switch in the built-in unit here in question, which represented the second step in the combined rheostat switch unit, he felt that the bracket type needed more attention and was more profitable.

Thus, from the fall of 1928 to the fall of 1929 he had done nothing on the 1928 work except to start a second sample, which he did not complete and which he mislaid or lost. In the fall of 1929 he life-tested the 1928 device, which he then mislaid or lost. From the fall of 1929 to the spring of 1930 he did nothing. In the spring of 1930 he ceased his activity completely and turned the job over to Haselwood. From the spring of 1930 to the spring of 1931 Haselwood did very little work on it, because other things were more important. Then, suddenly, Schellenger "pushed" the job to completion with added man-power, much overtime, concentrated experimentation, and manufacturing preparation. After almost three years of comparative inactivity, Schellenger reduced his idea to practice in less than four months of activity.

Modern industrial enterprise being what it is,^[6] Schellenger was fairly chargeable with knowledge that other inventors would be striving to discover the next improvement demand of the trade and that their inventive efforts were very likely to result in the same built-in unit which he had already conceived. Under these circumstances his long inactivity is inexcusable. He is certainly not entitled at this time, on any theory of fair play, to urge a superior right against these later inventors, who have expended their time, money, and energy in endeavors which would have been abandoned had Schellenger reduced his conception to practice within a reasonable time after 1928. Moreover, his sudden "push" in the spring of 1931, which is very inconsistent with his indolent conduct of 1929 and 1930, is indicative that the real reason was fear of Stoekle rather than the stimulus of reward for the diligent. This is further strengthened by the additional evidence that the barnacle type of switch, used by Stoekle, was dissected by Schellenger in April of 1931, which directly refuted Schellenger's testimony given at the interference proceeding.

To hold that Schellenger's conduct from 1928 to 1931 is proof of diligence is to reward indolence and encourage laxity on the part of the first conceiver, and to deter inventors in that their endeavors may prove worthless upon the discovery of any prior unembodied conception. Justice requires that the public reward only those who keep faith with it and who, having conceived ideas, reduce them as quickly as possible to practical public use. The public interest protects and rewards the first applicant for a patent, unless a more meritorious inventor has been hindered by circumstances beyond his control from conferring the invention on the public. The record in this case is clear to us: Stoekle is the more meritorious of the two.

Thus, taking all of the evidence together, it falls far short of establishing a priority of invention with that certainty which the law requires. We are satisfied that the probabilities are against Schellenger: uncertain or abandoned experimentation is not sufficient. Therefore, the interference deposition evidence, thought sufficient by the Board of Appeals in the United States Patent Office, can not be sustained in this court, after the crumbling blow dealt it by the additional evidence adduced by Stoekle in the trial court. The interference-judgment has been overcome by "testimony which in character and amount carries thorough conviction."

There is only one more contention by counsel that requires our consideration. Counsel contends that "Findings of fact by a trial court, so far as there is any testimony consistent therewith, must be treated as unassailable." Adamson v. Gilliland, 242 U.S. 350, 37 S.Ct. 169, 61 L.Ed. 356; U. S. v. Tyrakowski, 7 Cir., 50 F.2d 766; Ott v. Long Beach Co., 7 Cir., 70 F.2d 1; U. S. v. Beller, 7 Cir., 70 F.2d 463. We are not forgetful of the rule, announced in these cases, that compels our acceptance of the trial court's finding, when it is predicated on conflicting testimony of witnesses testifying in the trial court, and where credibility is largely determined by the personal observation of the trial judge. We are certain, however, that the rule does not apply here.

In the instant case the bulk of the record stands on deposition evidence. Findings of the trial court in such situations, while worthy of great consideration, are not conclusive in this court. In such cases the force of the story told by the witnesses and the corroboration of uncontradicted testimony is controlling, rather than the credibility of the story-teller. This court has spoken adequately on this point. Uihlein v. General Electric Co., 7 Cir., 47 F.2d 997, 1001-1006. The finding of the trial court in patent litigation may be successfully assailed, if the trial court clearly misapprehended or went against the clear weight of such evidence.

The decree is reversed, with directions to proceed in accordance with the views expressed in this opinion.

It is so ordered.

NOTES

[1] The inventors are the following: Schellenger; Stoekle; Meuer and Stevens; and Draving and Ingham. The defendant Chicago Telephone Supply Company owns the interest in defendant Schellenger's application for Letters Patent. The plaintiff-appellant Globe-Union Inc., owner of the stock of Central Radio Laboratories, is the assignee of the interest in the Stoekle application. Stoekle is represented by Lillian Stoekle, executrix of his estate, in this appeal. The defendant Cutler-Hammer Inc., owns the interest in defendants' Meuer's and Stevens' application. Draving and Ingham did not participate in this litigation.

[2] The issue in the Patent Office interference was declared in two counts, only the first one of which is involved in this suit.

[3] Draving and Ingham also filed an application, Serial No. 518,836, on February 27, 1931. They, however, did not take testimony, presented neither proof nor argument before the Examiner of Interferences in the United States Patent Office, and did not prosecute an appeal from his decision, whereupon the Commissioner of Patents dropped their application.

[4] Cutler-Hammer, Inc., Meuer and Stevens, the defendants-appellants, have appealed from that part of the decree dismissing their counterclaim against defendants-appellees. They allege that Stoekle is the prior inventor and that Meuer and Stevens antedate Schellenger. They pray priority only if this court finds that Stoekle is not the prior inventor; if this court finds with Stoekle, then their appeal need not be considered.

[5] The Stoekle unit was first reduced to practice on January 10, 1930, which is undisputed. The switch was improved by Cutler-Hammer in December of 1930 and manufactured by that firm exclusively for Stoekle's Central Radio Laboratories or Globe Union by March 11, 1931.

Meuer and Stevens, inventors for Cutler-Hammer, really invented this improved switch, but Globe Union had paid for the tools and for some of the development work and the switch was made under the Central Lab. (Stoekle) trademark. Because of this fact the switch was not generally known to the trade at this time.

Counsel for appellees urge that the switch used on the Stoekle device is not the invention of Stoekle but is instead the invention of Meuer and Stevens. We can not see the bearing of this on the question of Schellenger's priority.

[6] As in other fields of private endeavor, there is keen competition in the art to which these inventions belong. In the beginning, radio sets were equipped with separate volume controls and switches. The first step in combining these separate devices was taken in the outrigger type of combination unit, featuring the external bracketed switch and covered rheostat. The second step was taken in the built-in unit here in question, featuring the internal or barnacle switch and covered rheostat.