The suit is pursuant to Rev.Stat. § 4915 (1878), 35 U.S.C. § 63 (1940), to obtain letters patent. The only question is priority of invention, which in turn depends upon reduction to practice and diligence. In interference proceedings the Patent Office, through the Board of Appeals, awarded priority to Shepherd,1 although the Examiner of Interferences previously had found in favor of Abbott. The District Court also decided for Shepherd, on both diligence and reduction to practice, and accordingly dismissed Abbott’s complaint. He now appeals.
The invention relates to a process for weaving into cloth flexible yarn, composed wholly or in part of rubber. It is useful in making rubber garments or garments containing rubber strands. For instance, in making men’s clothing, the art has developed a fabric made with all worsted warp yarns, and mainly worsted filling or weft yarns interspersed every fourth, fifth or sixth yarn with an elastic rubber one. One of these elastic threads was known as “Lastex.”
Difficulties arose in weaving due to uneven tensioning of the elastic yarns. Variations in tension, caused by machine action, varied the stretch of the yarn. Such a variation in the finished cloth would draw in its edges unevenly, making them wavy, and would tend to make the cloth pucker or “cockle.” The gist of the invention, to meet these and other difficulties, is in coating the elastic thread with a sizing or liquid composition, so that when the coating hardens it forms a jacket around the thread and holds it in the tension which exists when the sizing is applied. The effect is to render the yarn sufficiently rigid to prevent the uneven tensions, but at the same time sufficiently flexible to permit weaving like ordinary yarn. The sizing is applied before the yarn is woven and is washed out or dissolved afterward. The yarn may be coated either in stretched or in relaxed condition as desired. One claimed advantage of interspersing rubber threads with worsted ones is to give the cloth greater capacity for holding crease when pressed. The process is useful also in weaving the more elastic garments women wear. Perhaps one element of genius in the invention is this achievement, at a single stroke, of greater rigidity for men’s apparel and greater elasticity for that of women.
Abbott conceived the invention about June 1, 1933. His application was not filed until September 22, 1934. Shepherd’s United States application is based on a British patent. His earliest date is March 19, 1934, when he filed a provisional specification in England, followed shortly by his complete one there and the parent United States application. His United States divisional application, which was the one involved in the Interference, was filed after September, 1934.
In view of his priority in conception, Abbott is entitled to priority in invention if he reduced it to practice prior to March 19, 1934, or if he maintained due diligence in perfecting it until the time of his application.2 The filing of Shepherd’s application constitutes a constructive reduction to practice as of March 19, 1934, if his disclosure is sufficient to enable one skilled in the art to practice the invention. At various stages in the Patent Office proceedings and in the District Court the issues have been: (1) whether Abbott reduced the invention to practice prior to March 19, 1934; (2) whether, if not, he was diligent in per*771fecting the -invention from a time prior to that date to September 22 following, when his application was filed; (3) whether Shepherd’s disclosure is sufficient to enable one skilled in the art to practice the invention, and therefore to constitute a constructive reduction to practice.
Except in respects not now important, material evidence was not greatly different in the Patent Office and the District Court. Notwithstanding this, as has been noted, there were contrariety of opinion within the trial tribunals and apparent confusion, if not inconsistency, in some of the parties’ contentions, especially concerning what is a sufficient reduction to practice and when this took place. Thus, Abbott claims conception and reduction to practice in 1933. But he admits and all of his evidence to show diligence is to the effect that he or Burke, through whom his experiments were performed, had not perfected the invention or found a satisfactory size before March 19, 1934, or in fact until late in May of that year. He contends, and must do so to retain priority in conception, that the gist of the invention is not in any specific sizing formula, but lies in the more general idea of applying any suitable sizing to the yarn, weaving it, and then washing out the sizing. Yet he urges strongly that Shepherd’s equally general disclosure, that is, apart from Abbott’s specific formulas, is not sufficient and therefore cannot be a constructive reduction to practice. Shepherd, on the other hand, seems to argue that his own disclosure is sufficient, and that Burke, who only devised formulas for Abbott’s sizing, not Abbott, is the inventor of the latter’s process. In view of this contrariety in result and reasoning, on the part of both the primary tribunals and the parties, the case is not one appropriate for disposal merely by applying the doctrine of Abbott v. Coe, 1939, 71 App.D.C. 195, 109 F.2d 449, and a statement of the evidence from which the various opinions have been drawn is necessary.
I
Abbott is an experienced and fertile inventor, gifted in designing machinery, especially for use in textile manufacturing, and i_i developing textile processes. He is also a business executive, engaged in the textile business since 1909 and in the textile machinery business since 1928. He is not an expert chemist or a specialist in the field of sizing yarns. He carries on his executive duties and his work of invention simultaneously, often conducting or directing several inventive projects at once. He is president of the Abbott Machine Company, which makes dyeing machinery among other things. In 1933 the company employed Mr. Burke, a trained chemist “primarily as an expert engineer, chemist, and dye man, to sell and engineer various problems in promotion of this dye machinery.” Burke lived in Mel-rose, Massachusetts, where he had a private laboratory in his home. His sales work required travel, so that he was away from home three or four nights each week.
Abbott testified he conceived the idea of sizing or coating rubber yarn for weaving purposes about June 1, 1933, and first mentioned it to Burke about July 4, just before leaving for his vacation. He returned just after Labor Day and gave Burke directions about following up the invention. These were that the best solution, apparently, would be “to stretch the yarn out and coat it with some material which would hold it in place; and then it could be woven or knitted and then the coating material washed off.” At the same time Abbott himself considered mechanical possibilities for controlling the stretch of the yarn in the loom or knitting machine, a problem distinct from the coating.3 Burke was told “to take almost any kind of rubber yarn that he could pick up and stretch it and coat it with various materials and see what we could do about a material which would hold it in stretched-out condition.” Among the possibilities Abbott mentioned were starch, casein and glue.
Burke tried out “some preliminary compounds” on ordinary small rubber bands. Before the end of September he reported to Abbott that a sizing would hold the strands extended. Following Abbott’s instructions, early in October Burke tried the coating on regular rubber yarn, samples of Lastex procured from a friend, Meade, who was an official of the United States Rubber Company, and made a brief memorandum stating, “Tried size on rubber thread; it holds . . . sets o. k. with spe*772cial size.” He reported this to Abbott soon afterward.
From then on Burke’s efforts were given to developing suitable formulas for sizing. He worked chiefly on three general types, based largely on dextrine, casein and thymol-iodoform. Eventually he developed three specific formulas, which he termed the “dextrine type,” the “casein type” and the “thymol-iodoform type,” respectively. They were incorporated in that order of preference in Abbott’s application as “preferred compositions of size.” But the specification stated: “The size may be any suitable adhesive composition of sufficient tenacity for the purpose indicated, but not destructive of the flexibility necessary for winding and weaving, and capable of subsequent removal, preferably by water.” Formula No. 1, the “dextrine type,” was the most satisfactory and the one which finally convinced Abbott the search for the right compound had been concluded successfully. Just when this was developed is not entirely clear. But it was on May 25, 1934, when Burke took coated specimens of the Lastex yarn to him, that Abbott approved the results finally and told Burke to write up a report covering his whole work. Burke did so, and on May 29 Abbott instructed his attorneys to proceed with the preliminary search and a patent application.
The evidence for Abbott consisted largely of his own and Burke’s testimony and a few documentary and physical exhibits, although Mrs. Burke gave supporting evidence, and one of Abbott’s attorneys gave evidence to show diligence from May 29 to September 22, 1934. According to the testimony of Abbott and Burke, the latter worked continuously on experiments to develop a suitable sizing from September, 1933, to the latter part of May, 1934. Apart from procuring the yarns and reporting to Abbott, the work was done in the laboratory in Burke’s home. This was in the evenings three or four days a week or over week ends when he could be at home. The remainder of his time was spent in travel on his duties as sales and engineering consultant.
Burke saw Abbott usually once a week on Fridays at the latter’s home in New Hampshire. Then he would report on his work for the company and also from time to time on his progress with the formulas. During these conferences Abbott instructed him concerning further procedure. Their testimony is that from the “fall of 1933” to April, 1934, Burke .made thirty or forty experiments on the Lastex yarn he procured in October from Meade with different formulas, and that it took several days to complete any one experiment with any one formula. The work included compounding the formula, putting it in solution, stretching, coating and drying the yarn, testing it, and removing the sizing. In April, 1934, Abbott rocured a new supply of Lastex from Adamson Bros, in New York and the experiments thereafter continued with this yarn.
Burke early found that the “casein” and “thymol-iodoform” formulas offered less favorable possibilities than the “dextrine” type, because of greater difficulty in procuring the materials and in dissolving or washing out after weaving. He therefore concentrated on the dextrine formula. Most of his work was done on that, extending apparently from October, 1933, to May, 1934.
From the beginning the formula was satisfactory as to rigidity, and apparently also as to washing out. But its flexibility for withstanding the operations of the weaving and knitting machines did not quite satisfy Abbott. This was the problem on which Burke worked exclusive’^ for nearly nine months. When Abbott procured the new Lastex in April, he turned some of it over to Burke and, according to the latter, “His instructions to me then were that he felt that the formula that I had showed to him previously, that is the results of it were so close to what he wanted that I should concentrate all my efforts on just that one thing and not take any departures whatever, and to modify that by various experiments, to overcome that inflexibility and scaling of the formula * * Burke then modified it by removing the small amounts of glue and glycerine it had contained and substituting glucose. This took “six or seven weeks of further work,” and “at least six or seven” further experiments, during April and May.
Documentary and other exhibits introduced to support this version of extensive and long-continued chemical experimentation upon the narrow phase of the formula’s flexibility included: a strand of the original rubber bands sized by Burke in September, 1933; two detached loose-leaf notebook pages containing brief memoranda, one dated October 5, 1933, signed by Burke, and witnessed by his wife on October 7, summarizing in a few lines *773what he had done up to that time and containing later notations, one dated “Apr. 1934,” that “E. J. A. handed bbn rubber yarn cotton coated — will size See report Mr. E. J. A. June 1934 Report Dec. 1934,” the other sheet containing brief chemical notations, apparently for the three types of formula, not dated; a rough pencil sketch of a projected sizing machine, dated “Dec. 1933” and signed by Burke; and a printed form of acknowledgment of an order for ten pounds of Lastex, dated March 17, 1934, sent by Adamson Bros. Company to Abbott Worsted Mills, Wilton, N. H.; a notice of shipment of this order dated March 21; and a letter from Abbott to Adamson Bros, dated March 20, which appears to be concerned almost entirely with business matters, though possibly it had slight relation to the invention.4 The record also includes a copy of Burke’s final report to Abbott, several pages long, made in June, 1934. Exhibit 6-A was a deteriorated sample of Adamson Brothers’ shipment of Lastex, and Exhibits 38-A and 38-B were Burke’s final specimens of Lastex which Abbott found satisfactory.
In summary, the physical exhibits covering nine months’ work comprised samples of the original coated rubber bands (September, 1933); a sample from the Adam-son shipment of March, 1934; and specimens of the final coated yarn (May, 1934). The documentary evidence for the same period includes Burke’s two brief loose-leaf sheets, dated October S, 1933, with their later brief notations; his rough pencil sketch of the projected sizing machine dated “Dec. 1933”; Adamson’s acknowledgment of Abbott’s order for Lastex, of March 17, 1934; the notice of shipment of March 21; Abbott’s letter to them of March 20; and Burke’s final report of June, 1934.
II
If Abbott’s evidence is taken fully at face value, he not only conceived the invention about June 1, 1933, but he and Burke were at work upon it constantly and continuously from early in September of that year to the end of May, 1934. From that time forward his attorneys were engaged in preparation of the application for filing, and it seems not to be questioned that there was due diligence in this phase of the work from May 29 to September 22, 1934, when the application was filed.
Shepherd, however, attacks the evidence given on Abbott’s behalf, both for want of legal sufficiency to show diligence and reduction to practice prior to March 19, 1934, and for want of credibility. He divides Abbott’s activities into three periods: first, from conception in June, 1933, to an indefinite time in December, 1933; second, from the December date to an indefinite one in April, 1934 (during which Shepherd entered the field); and third, from the April date to September 22, 1934. Discounting the seriousness of the work done in the first period, he says the second is barren of any credible evidence of continued and diligent activity, as is the third up to May 29. This is especially true, he argues, of the time extending from a date prior to March 19, 1934, when Shepherd filed, to May 29 that year, a period which appears to be conceded, and the trial court found, is crucial on the question of diligence. Shepherd also says that Abbott did not reduce the invention to practice prior to March 19, 1934, apparently because his work on the formula was not finished and because the process was never tried out in actual weaving or knitting. Abbott, as has been said, relied not only on his own efforts, but also upon the asserted insufficiency of Shepherd’s disclosure, basing this apparently upon the absence of a specific formula.
In this state of the evidence and the issues the trial court made findings of fact and conclusions of law in Shepherd’s favor, too long for inclusion here, but material parts of which were as follows:
Findings of Fact
“12. Both of the Patent Office tribunals held and this Court finds the defendant Thomas Lewis Shepherd to be entitled to a date of conception and constructive re*774duction to practice of the invention in issue as of March 19, 1934, this being the date of filing of Shepherd’s British application which, like his later filed United States applications and his said patent, embodies adequate disclosures of the invention in issue.
“13. Upon the evidence adduced on behalf of Abbott in the Patent Office interference, the Board of Appeals found that none of the plaintiffs’ activities could be accepted as constituting a reduction to practice of the invention prior to Shepherd’s date of March 19, 1934; also, that Abbott had failed to prove diligence in reducing the invention to practice, especially throughout the period from March 21, 1934 until May 29, 1934.
“14. The evidence presented before this Court consists of the Patent Office Record with certain additional matter. So far as physical evidence is concerned, the sole addition to the Patent Office Record is an exhibit comprising some samples of sized rubber yarn said to have been newly discovered. The plaintiff Abbott and his assistant Burke, both of whom had testified in the Patent Office proceeding, gave further oral testimony before the Court and were cross-examined at length. One new witness (Burke’s wife) testified by deposition in support of some of the early work claimed to have been done by Burke on Abbott’s behalf, and an attorney testified that he made a novelty search for Abbott (long after Shepherd’s date).
“IS. Abbott had not reduced the invention to practice prior to Shepherd’s date of March 19, 1934. He never made the fabric nor a sufficient amount of sized yarn to weave in fabric, nor tested it in weaving or knitting, although the facilities for doing all these things were readily available to him.”
“17. The plaintiffs’ evidence fails to establish diligence, regardless of the purpose of the alleged activities. This is especially true of the period between March 21, 1934 and May 29, 1934, to which the new evidence was in part directed.
“18. The testimony that many experiments were made for determining a satisfactory formula for the size was not convincing either as to the time when made or the need for them.”
Conclusions of Law
“4. The plaintiffs have failed to sustain the burden of proving by evidence which in character and amount carries thorough conviction, that the Board of Appeals erred in holding Shepherd to be the prior inventor and entitled to a patent for the invention in issue.”
On these findings and conclusions, the court entered the following order:
“So far as any question - of reduction to practice of the invention involved in this case is concerned, I am satisfied that Abbott had not reduced it to practice before March 19, 1934. He had never made the fabric, nor a sufficient amount of sized yarn to weave a fabric, nor tested it in weaving or knitting.
“The evidence of the plaintiff as to diligence is largely based upon memory; the documentary evidence consisted in part of memoranda and letters, which are connected by oral testimony with the facts in issue. When it is borne in mind that the plaintiff Abbott had had experience in obtaining patents and his employee Burke was a research chemist it is difficult to understand why more accurate entries were not made showing the progress of the work. I am not satisfied that the plaintiffs have maintained the burden of proof either of reduction to practice or of diligence. As to the latter this is especially true of the period between March 21, 1934 and May 29, 1934. I think too that suspicion always attaches to testimony introducing memoranda which it is claimed were not discovered until after a former hearing.
“I think that the plaintiffs have failed to sustain the allegations of the complaint, and that it should be dismissed with costs.”
Ill
In our view of the case it is necessary first to clarify the apparent confusion relating to what constitutes the invention or, more precisely, to the consequences for reduction to practice and diligence which grow out of that question. If the gist of the invention is in a specific formula or formulas for sizing, obviously Shepherd’s disclosure is not sufficient and he has never conceived the invention. It would follow also that Abbott has conceived it, fully disclosed it, reduced it to practice, at least constructively by the filing of his application, perhaps also by completing his experiments, and that he is entitled to the patent, whether or not this work was done diligently.
On the other hand, if the invention consists in the general idea of coating elastic *775yarns with sizing or other material to make them rigid, yet sufficiently flexible for weaving purposes, weaving the yarn into cloth in its sized condition, and washing out or otherwise dissolving the sizing after the cloth is finished, that also has obvious consequences, generally of a contrary character, for the issues and the parties. This view of the invention necessarily affects the questions, what are a sufficient disclosure, a constructive and an actual reduction to practice, and due diligence. They will be determined differently, depending upon whether the rather simple general idea or the more complex problem of the specific chemical formula embodies the invention.
It is too late to take the view that the invention lies in devising a specific sizing formula or formulas. The interference proceedings, as well as the previous ones by both parties in the Patent Office, and the trial were conducted throughout on the basis that the invention is embraced in the general idea. Both parties concede this, though each has departed for argumentative purposes from the necessary consequences of that fact at one stage or another of the proceedings. Thus, Abbott states in his brief: “The gist of the invention is the coating of an elastic thread with a sizing composition in liquid form, the thread being either in stretched condition or relaxed condition as desired, so that when the coating hardens it forms a jacket around the thread and holds it in its then condition. The thread will therefore neither extend nor contract in the fabric-making operations. * * * Thereafter the coating is removed * * * preferably made of a sizing soluble in water or in soap and water so as to be removed during the usual finishing operations on the cloth.” His specification says: “The size may be of any suitable adhesive composition of sufficient tenacity for the purpose indicated, but not destructive of the flexibility necessary for winding and weaving, and capable of subsequent removal, preferably by water.” (Italics supplied) It then sets forth as “preferred compositions of size” the three specific formulas. But Abbott’s most significant statement is in a letter to his attorneys dated July 3, 1934: “7 do not think we need to go into■ the composition of the size at this time. We could merely describe it as a size made of starch, glucose, etc. making this application purely one on the process. Later on, when we get some formulas which may present something of novelty, we can make further application, if need be, on the composition of the size.” (Italics supplied) The attorneys responded on August 30: “We are not sure that it is safe to describe the size in the very general language employed in this draft of the application, for the reasons that the Patent Office or a Court might hold that you had not adequately described a size suitable for use in the process. This would depend somewhat upon the question whether most ordinary sizes used for textile purposes would work satisfactorily in your process.” 5 (Italics supplied) The letter went on to point out that if only certain sizes ordinarily used would work, there was “some danger” of the application being rejected for insufficient disclosure or, if granted, later held invalid for this reason, and suggested inserting “a brief description of one size which will work, by inserting after the general description of the size contained in the application the words, ‘for example’ followed by a specification of one suitable size.” (Italics supplied) On September 5, Abbott replied: “We agree with you that it would be in order to define the particular kind of size used, but in a definite description.” Two days later he sent the three formulas to the attorneys, which they inserted in the application in the form stated above. None of Abbott’s original fifteen claims includes a specific sizing formula, but all use general terms such as “applying size,” “applying and hardening soluble substance thereon,” “fixed by soluble adhesive,” “including a soluble size,” “including initially suppressing the inherent elasticity of the elastic yarns.” Finally, the five counts set up in the interference proceeding were cast in equally general terms, “applying size thereto” or “to the yarn,” “having a soluble size deposited thereupon,” none specifying any particular formula or composition.
IV
Notwithstanding all this, Abbott still contends that Shepherd’s disclosure of the invention is not sufficient to enable one skilled in the art to operate it, and therefore his filing of March 19, 1934, was not a constructive reduction to practice, a posi*776tion apparently contrary to the one he took in the interference proceedings.6
Shepherd’s provisional British application, which entitles him to the date of-March 19, 1934, sets forth the difficulties of controlling the stretch of elastic or rubber threads caused by weaving and that the thread itself may vary in flexibility. It then states the object of the invention is “to overcome this loss of control by coating the yarn, either in its natural relaxed condition or having a desired degree of stretch, with a suitable kind of soluble adhesive, such as glue, solutions of cellulose or the like, or a suitable varnish.” Application by coating, spraying or otherwise is specified, so as to impart a rigid or semirigid “set.” “The adhesive used,” it is further stated, “is preferably of such character that it will quickly dry or any artificial drying means may be used. * * * The thread or yarn having thus been coated * * * is then woven or knitted into a fabric. * * * Subsequently by a later process the adhesive coating is removed or otherwise eliminated by washing, dissolving or otherwise leaving the fabric with its original flexibility, or with the flexibility it should possess.”
Shepherd’s disclosure in his United States divisional application, so far as is material, was: “The object of the invention is to so control the thread before and during weaving that it exerts no tension, that is to say, it is in a static condition, so as to thereby eliminate all irregular patches, lines or zones in the fabric. Another object * * * is to temporarily deprive the rubber thread of its elasticity to enable it to be conveniently worked, the elasticity being finally restored to it again, by and during the subsequent dressing, finishing and other operations necessary to completely finish the fabric. * * * In carrying the invention into effect the rubber thread a [referring to a drawing] is prepared in any suitable way. * * * A thread thus prepared may then be extended for weaving, knitting, lace making or any other purpose * * * and the thread a is then coated either in its natural relaxed condition or under any desired degree of stretch as above described, with a suitable kind of soluble adhesive as shown at 11 in the drawing. Such an adhesive coating which should be of a quick drying character may consist of glue, gelatine, a solution of cellulose, starch, wax, resin or the like or a suitable varnish. It may be applied by coating, spraying, painting or otherwise and the coating is then allowed to dry to such a degree as will impart to the thread a rigid or semi-rigid set so that the resulting thread will not possess the characteristic of extensibility but is deprived of its elasticity.” Then are described the succeeding steps of weaving or knitting and removal of the sizing “by being boiled or heat treated in any suitable way.” One of Shepherd’s claims is even more specific than Abbott’s as to the nature of the sizing, namely, claim 2, which specifies “coating an elastic rubber thread with soluble material such as gelatine,” though the others use more general language, “coated with soluble material,” or “a suitable soluble material.” It will be noted that the complete specification adds gelatine, starch, wax and resin “or the like,” as “suitable adhesives,” to the “glue, solutions of cellulose or the like, or a suitable varnish” mentioned in the provisional one.
The finding was clearly right that Shepherd’s disclosure was sufficient and establishes a constructive reduction to practice as of March 19, 1934. In this the Examiner of Interferences, the Board of Appeals and the District Court are agreed. The finding is correct first because the invention consists in the general idea, not in the specific formula or formulas for sizing. Shepherd clearly sets forth that idea. Further, it is one which, on the record, must be taken as capable of being made to work by one skilled in the art.
Abbott makes the contrary argument as follows, "No single substance such as the Shepherd specifications prescribe would do the work,” and relies upon Burke’s “uncontradicted testimony” to that effect. (Italics supplied) The argument and the testimony attempt to pin Shepherd down to what his disclosure does not require and what in fact Abbott himself does not require. This is use of each of the substances mentioned by way of example, singly and exclusively, without modification or combination in any degree with any other. Shepherd does not limit the *777coating to the substances he specifically mentions, nor does he require that they be used singly or in unmodified purity. The language of the provisional British specification is: “a suitable kind of soluble adhesive, such as glue, solutions of cellulose or the like, or a suitable varnish.” The coating is one which will impart to the fabric “a rigid or semi-rigid ‘sef, so that the resulting fabric or thread will be rendered inflexible or semi-flexible,” and such as, subsequent to weaving or knitting, may be “removed or otherwise eliminated by washing, dissolving or otherwise leaving the fabric with its original flexibility, or with the flexibility it should possess.” (Italics supplied) The complete specification requires “a suitable kind of soluble adhesive,” which should be “quick drying,” but “if necessary the drying can be accelerated by artificial drying means of any kind * * * may consist of glue, gelatine, solution of cellulose, starch, wax, resin or the like, or a suitable varnish * * * may be applied by coating, spraying or otherwise, * * * to such a degree as will impart to the thread a rigid or semi-rigid set,” etc.
Both specifications obviously were drawn broadly with a view to the existing art in sizing, as well as in other respects, and were intended to leave to the artisan the selection of materials and methods known to the art to accomplish the specified results.7 They designedly avoid limiting the process to particular or specified materials, methods of application and removal, or exact degrees of consistency or flexibility. Those mentioned are by way of illustration, example or general guidepost. They presuppose adaptability to different kinds of threads and fabrics.8 The general characteristics of solubility, quick drying, rigid or semirigid set sufficient for holding the yarn without stretch in the loom or knitting machine are required. But room is allowed for play of the artisan’s skill in selecting sizings which meet these conditions in particular machines and fabrics.
The reason for making the disclosure in these broad terms is obvious. It is that little or no protection would be afforded the inventor if the specification were limited to a single specific formula or to a few set forth with 'chemical accuracy. Such a patent would be almost worthless. That undoubtedly is why Abbott was unwilling, first, to specify any particular formula at all, second, to limit his disclosure to the three formulas he finally included on the advice of his attorneys. He wanted a patent which would cover his invention and do so adequately. Only one broad enough to cover the general idea could do this. One limited to his specific formulas could give him monopoly of only a very narrow segment of the field, leaving open to his competitors or the public by far the larger area of his conception. This he avoided by specifying the formulas as merely “preferred compositions of size.” He too left room for the artisan’s skill in his general and controlling specification quoted above. This, like Shepherd, told the artisan broadly to use his art in selecting the composition to accomplish the purpose specified. That Abbott added “pre*778ferred compositions” also to show specific methods of doing this, only one of which according to all of the evidence finally and completely satisfied his meticulous standards for detail, in no way nullifies the freedom he gave the artisan to apply others. Thus, Abbott’s application by its very terms recognizes that the selection of specific materials and methods of application and removal are matters within the scope of the existing art. His letter of July 3, 1934, to his attorneys makes this more than conclusive. Language could not be stronger to annihilate his own argument that Shepherd’s disclosure is not sufficient. The latter is entitled to the date of March 19, 1934, as one of constructive reduction to practice.
It may be added that, though Burke testified the particular substances mentioned by Shepherd, used separately, singly and unmodifiedly, would not do the work, neither he nor anyone else testified that no other compositions than Abbott’s formulas would accomplish it. There was no evidence that successful compositions are not within the compass of the existing art or cannot be devised by persons skilled in it. The so-called “uncontradicted testimony” of Burke was therefore beside the point or, more accurately, wide of the mark.
V
It remains to consider Abbott’s reduction to practice and diligence. From what has already been said, it is obvious that most of his work through Burke, namely, from some time in October, 1933, to May 25, 1934, was not necessary to reduce the invention to practice. The opposite might be true, if the invention had consisted in developing specific sizing formulas, except for one fact. That is that Burke developed his three general types of formula very early in his work, apparently in October, 1933. He found that all three would work experimentally. None was tried in actual weaving or knitting. But almost as soon as he found the general formulas, he stopped work on two of them, and concentrated thereafter upon refining the feature of flexibility in the dextrine type. This continued for nine months. But when the application was drawn and filed Abbott not only stamped all the formulas merely as “preferred compositions.” He included the two upon which work had been finished the preceding October.
The significance of this is obvious. The casein and thymol-iodoform types are included as workable sizings. They were developed almost a year before Abbott’s application was filed, nine months before he pronounced the dextrine type satisfactory. In other words, the invention as such was completed in the fall of 1933. Thereafter there was refinement in a detail of one form of its application. This however was not essential to workability if, as Abbott’s application necessarily implies, the other two formulas are workable. In short, if his evidence is accepted fully at face value, Abbott completed the invention in the fall of 1933, probably in October, but waited about nine months before instructing his attorneys to prepare and file his papers. During this time he was occupied busily with a nonessential refinement of one form of the invention’s application, which he himself characterized in the letter of July 3, 1934, as not presenting anything of novelty.
One conclusion which might be drawn from these facts is that Abbott reduced his invention to practice in 1933. The Acting Examiner of Interferences made no ruling on this, since he regarded Abbott as having used reasonable diligence from just prior to March 19, 1934, to the date of his application. The Board of Appeals found the opposite on diligence, holding that Burke’s work in refining the dextrine formula was not relevant to the counts of the Interference. The opinion states, “it appears that the particular size involved in this method is not critical” and there was “no indication of diligence from March 21, 1934 until May 29, 1934.” The Board also observed “that it appears conclusive that none of Abbott [sic] or Burke’s activities can be accepted as constituting a reduction to practice prior to Shepherd’s record date of March 19, 1934.” This followed statements that Abbott had unlimited opportunity to proceed more directly and diligently “in that he was president of, and apparently a large owner of interests in extensive textile industries” and that “it would be a matter at most of only a slight effort to size up and run through a loom some of the ‘Lastex’ yarns.” The District Court’s similar finding, set forth above, is based on like reasons. Although a contrary one might be sustained, in view of the invention’s relatively simple nature and the testimony that actual demonstration was not necessary to confirm the experimental one, we cannot say that the finding made both by the Board of Appeals and by the District Court *779was not supported by sufficient evidence or that it constituted an abuse of discretion. Accordingly the finding that Abbott had not reduced the invention to practice prior to March 19, 1934, must be sustained.
But if that finding were to the contrary, the record shows conclusively that the Board and the Court were right in finding that Abbott did not exercise diligence. The Board’s finding apparently was based solely on the ground that the evidence was legally insufficient for this purpose. The trial court’s findings, numbers 17 and 18 set forth above, appear to be based both on this ground and on want of credibility in the evidence presented for Abbott. Accordingly it concluded that Abbott had “failed to sustain the burden of proving by evidence which in character and amount carries thorough conviction, that the Board of Appeals erred in holding Shepherd to be the prior inventor and entitled to a patent for the invention in issue.”
Credibility is largely for the trial court’s determination. In this case the record is not such that we can say there was error in refusing to give full credence to the evidence favorable to Abbott. When account is taken of the comparatively simple character of the invention, its large reliance upon the prior art, the ease with which the basic problem was solved, and the early date at which this occurred, the evidence of long and exhaustive experiments upon a mere refinement of one phase of one formula for use in only one of many possible applications well might strain credulity. The strain is not lessened by the paucity of physical exhibits and laboratory notes or memoranda made during this long research. Research chemists are notably meticulous in this respect, especially when engaged in extensive and baffling projects. The character of the memoranda, especially in content and dating, as well as the volume, add to the doubt. In these respects the cases Abbott cites, upholding the credibility of testimony based largely on memory, are distinguishable.9 The evidence with which they were concerned did not involve such inherent marks of untrustworthiness. Nor were the arguments drawn from it so wholly inconsistent as some presented here. On this basis alone, therefore, the finding concerning diligence could not be set aside.
But the case for Abbott would hardly be better if full credence had been given to all the evidence admitted on his behalf. The reasons for this have been stated. The principal one is that, after completing his invention, he delayed too long in refinement of a minor and nonessential detail of one application of his process, work which belonged to the artisan, not to the inventor. However necessary and convincing that activity might have been if the invention claimed had been in the specific formula it finally developed, in relation to the invention in issue it was at most an artisan’s sidetrack where the inventor had no business to be when others were coming along the main line. Perfection of utility is to be encouraged.10 But delay, while one is engaged only in what is already known to the art after the essential idea has been proved, is not that diligence which is required to secure priority in invention.
The judgment is affirmed.
The parties to the suit, including the corporate assignees, will be designated by the names of the individual parties, who claim respectively to be the inventor.
Cf. Woods v. Poor (1), 1907, 29 App.D.C. 397; Kitchen v. Smith, 1913, 39 App.D.C. 500; Saurer v. Groebli, 1916, 45 App.D.C. 298; Dickinson v. Swinehart, 1920, 49 App.D.C. 222, 263 F. 474; Gammeter v. Backdahl, 1920, 50 App.D.C. 45, 267 F. 347; Joy v. Morgan, 1924, 54 App.D.C. 110, 295 F. 931.
The opinion will not be burdened with the details concerning these and other projects on which Abbott was working, although Shepherd contends that some of the documentary and other evidence, which Abbott says relates to the invention now in issue in fact relates to these other projects.
The letter expressed thanks for “the opportunity of talking with your Mr. Percy and Mr. Seth Adamson in regard to the use of Lastex yarn in our fine fancy men’s wear fabrics,” explaining Abbott’s textile and textile machinery manufacturing facilities, the merits of his organization for adapting its equipment and doing necessary experimental work “to make these fabrics right,” and expressing appreciation for “the information you gave me as to how to handle this yarn and worsted fabric.” It also notes their acknowledgment of order “covering shipment of sample lots,” and ends: “As soon as these are received, we will start work with them.”
Cf. note 7 infra and text.
The opinion of the Acting Examiner of Interferences states: “Abbott does not contend that the filing of this provisional specification does not constitute a construetive reduction to practice by Shepherd, but urges that it is overcome by his own priority proofs.”
During the interlocutory stage of the Interference, Abbott moved to dissolve on the ground that Shepherd had no right to make the count. In denying the motion, the Primary Examiner ruled:
“It is held that in view of the well known use of sizes in the textile arts that the skilled artisan provided with the teachings of the Shepherd application could readily formulate the specific size required to produce the desired result without entering the realm of the inventor.”
Commenting upon this, the Acting Examiner of Interferences said:
“In view of this holding in the case, it seems proper to accept the statement made on Shepherd’s behalf before the Primary Examiner, ‘that it is quite superfluous to refer specifically to the solutions suitable for rubber thread when bare or when covered’ and it is obvious that if such details were not needed in Shepherd’s disclosure, they correspondingly would not be needed in Abbott’s.”
Accordingly it was held in Abbott’s favor and against Shepherd’s contention that Burke, not Abbott, was the inventor, that Abbott had established conception in (June) 1933, a ruling obviously impossible if devising a specific sizing formula were not the work of the skilled artisan rather than invention. It may be noted again that these rulings coincide completely with Abbott’s own views, as expressed in his letter to his attorneys of July 3, 1934.
The opinion of the Board of Appeals states: “Each count [in the Interference] definitely includes some limitation to the weaving or knitting yam.”
Storck v. Reichhelm, 1916, 44 App.D.C. 438; McAfee v. Gray, 1918, 47 App.D.C. 237; Lautenschlager v. Glass (1), 1918, 47 App.D.C. 443; Browning v. Johnson, 1921, 50 App.D.C. 335, 271 F. 1017; Bijur v. Bendix, 1923, 52 App. D.C. 240, 285 F. 974; Penn Oil Co. v. Vacuum Oil Co., 1931, 60 App.D.C. 96, 48 F.2d 1008. See also the authorities cited in note 2 supra.
See note 2 supra.