STOODY CO. v. MILLS ALLOYS, Inc., et al., 67 F.2d 807

SAWTELLE, Circuit Judge. This suit was brought by the appellant for the alleged infringement by the appellees of the appellant’s United States letters patent No. 1,757,601, issued May 6,1930, and covering a welding rod used for applying welding on facing for well drilling tools and the like, which are subject to extreme abrasion. The welding rod in its preferred form consists of a mild steel tube of suitable length, closed at both ends, and filled with pieces of tungsten carbide. In use, the steel of the tube is melted by the heat of an acetylene torch, and is fused to the steel of the bit.

The tungsten carbide particles apparently are not affected by the heat of the torch: they neither lose their hardness nor do they dissolve in the molten mild steel. They are distributed through the deposited steel of the tube, which, when it solidifies on the bit, forms a matrix for the evenly distributed tungsten carbide particles, usually of a size ranging from a pin head to a match head. Because of its hardness, which closely approaches that of a diamond, and which is retained through the welding operation, tungsten carbide resists the abrasion to which the bit is subjected in drilling.

This prolongs the life of the bit, which, when faced with this rod, wears as follows: The mild steel matrix in which the tungsten carbide particles are imbedded wears away between those particles, exposing them in the form of small projecting hard fragments, which perform most of the cutting action. The patent was issued on an application filed on January 30, 1928, by Winston F. Stoody, Shelley M. Stoody, and Norman W. Cole, who assigned their joint application to the appellant. The patent was issued to the appellant.

Under a prior stipulation, an uncertified copy of patent No. 1,757,601 was introduced in evidence at the hearing before the master. On motion of the appellant, an order of reference to a special master was made by the District C'ourt. The appellees agreed to such reference.

The order empowered the master “to make his conclusions of fact in issue and recommend the judgment to be entered thereon; * * * to do all things and to make such orders as may be required to accomplish a full hearing on all matters of fact and law in issue, * * * as fully and completely as though this reference had not been made and as though this cause had been tried before the Court, reserving to the Court the full right and power to review and determine all questions of fact and law,” etc.

The master found that the appellees’ “article of manufacture and sale differs in no material detail from the device of the patent,” and that it is “intended for use in the same manner.” It seems clear that, if tho appellant’s patent is valid, the appellees’ welding rod is an infringement of it. Our own independent examination of the two rival devices, which are part of the exhibits before us, has convinced us of their virtual identity. The defenses relied upon by the appellees, as enumerated in the briefs and in the master’s report, were as follows:

(1) There is no invention disclosed; (2) the patent was obtained by fraud; (3) the invention was not joint in nature; (4) the subject-matter of the patent was first invented by the appellee Mills. The master found in favor of the appellees on the first defense — -lack of invention on the part of the appellant’s welding tube— and found against the appellees on the other three defenses. Accordingly, he recommended that a decree be entered finding the appellant’s letters patent invalid. Both sides filed exceptions to the master’s report, and all exceptions were by the court overruled and disallowed.

The court below entered a final decree adopting the master’s findings of fact and conclusions of law, and adjudging the appellant’s letters patent to be invalid. From that decree the present appeal was taken. Under our view of the ease, it is necessary to consider only the defense in which the appellee’s position was sustained by the master and the court below; namely, that the appellant’s patent was invalid for want of invention.

In a brief memorandum opinion handed down a few days before the final decree was recorded, the court below expressed the view that “the grant of the patent to plaintiffs [appellant] invests it with no presumption of validity and plaintiff has failed to sustain any claim of invention by a preponderance of the evidence.” This holding by the court is made the subject of three assignments of error. In this connection, we may observe that error is not assignable to the opinion of a court. This court has repeatedly so held. In Yangtsze Rapid S. S. Co. v.

Deutsch-Asia^ tisehe Bank, 59 F.(2d) 8,12, we said: “But, while we do not agree with the reasoning of the lower court, we do concur with the conclusion reached by it. It is not necessary that a judgment be affirmed for the precise reasons that seemed controlling in the lower court. In McCloskey v. Pacific Coast Co. [C. C. A.] 169 F. 794, 801, 22 L. R. A. (N.

S.) 673, the late Judge Gilbert of this court said: ‘But notwithstanding that the theory upon which the court below awarded its injunction may have been erroneous, the injunction must not be disturbed if in the pleadings and the proofs we may discover any tenable ground upon which it may be sustained.’ [Cases cited.]” Similarly, in Stoffregen v. Moore (C. C. A. 8) 271 F. 680, 681, the court said: “The assumptions of fact complained of in assignments of error 1 and 2 are found, if anywhere, in a memorandum opinion of the trial court incorporated for some reason in what is called a bill of exceptions.

These two assignments of error present nothing for review: First, because they are based upon the opinion of the court, which cannot be the basis of an assignment of error. The opinion may be' wrong, and still the judgment be right. Second, we are by statute forbidden from reversing a judgment for error of fact on writ of error. Rev. St. § 1011; Comp. Stat. § 1672 [28 USCA § 879].” That this rule applies to asserted errors of law as well as .to those of fact, appearing in an opinion of a court, is clear from the ease of United States v. Porter Fuel Co. (C. C.

A. 8) 247 F. 769, 770, 771, which is on all fours, on this point, with the instant controversy. There, too, the validity of a patent was involved, and alleged error on the part of the lower court as to the question of burden of proof was involved. The appellate court said: “When counsel for plaintiffs come to demonstrate why the court erred in dismissing the complaint they present their argument under two headings as follows:

(1) The District Court erred in its ruling in respect to the burden of proof. * * * These headings are practically new assignments of error, and are based upon what the trial court said in delivering its oral opinion. The opinion of the court was not the subject of exception or assignment “of error. The* reasons given in the opinion for the judgment of the court might be wrong, and still its judgment right. It is what the court did, and not what it said, which is subject to exception and review.” See, also, Smart v. Wright (C. C. A. 8) 227 F. 84, 85; Nowata County Gas Co. v. Henry Oil Co. (C. C.

A. 8) 269 F. 742, 744; Wineinger v. Union Pac. R. Co. (C. C. A. 8 ) 276 F 65, 67; City of St. Paid v. Certain Lands in City of St. Paul, Minn. (C. C. A. 8) 48 F.(2d) 805., 807. Nevertheless, because the question of burden of proof, the presumption of validity attaching to a patent, and kindred problems relating to the quantum of proof, may be important in evaluating the master’s report and the concurring decree of the District Court, we will consider the above three assignments of error on their merits.

In its brief, the appellant, in connection with the three assignments in question, says: “The general rule is that a patent is presumptively valid. This presumption arises from the grant of the patent by the Patent Office after the application has been examined thoroughly by the Examiner.” The foregoing excerpt unquestionably correctly states both the rule of law and the reason therefor. In Wilson & Willard Mfg. Co. v.

Bole, 227 F. 607, 609, the late Judge Rudkin, of this court, said: “The general rule that a person who attacks the validity of a patent issued to another must make out his case by clear and satisfactory proof, or by proof beyond a reasonable doubt, will not be gainsaid.” Again, in Bankers’ Utilities Co. v. Pacific Nat.

Bank, 18 F.(2d) 16, 18, the late Judge Dietrich, also of this codrt, observed: “In their position plaintiffs are fortified by the presumptions attending a patent * * * and by the fact that their device is a commercial success and has brought on imitation. [Many eases cited on each point.]” Still another expression on the subject by this court is to be found in H. J. Heinz Co. v.

Cohn, 207 F. 547, 554: “The patent is always presumptive evidence of prior invention and novelty and, when produced, makes for the patentee or his assignees á prima facia ckse which, to overcome, requires proof of the lack of novelty, and that some one else is the prior inventor, beyond a reasonable doubt. The authorities are uniform to this purpose. Coffin v. Ogden, 18 Wall. 120, 124, 21 L. Ed. 821; Cantrell v. Wallick, 117 U. S. 689, 695, 6 S. Ct. 970, 29 L. Ed. 1017; Parker v. Stebler, 177 F. 210, 101 C. C. A. 380.” See, also, Detroit Motor Appliance Co. v. Burke (D. C.) 4 F.(2d) 118, 122; 48 C.

J. 181, and 354r-355, §§ 262 and 574; Walker on Patents, infra, § 116, p. 141. In the instant case, however, a special fact situation is presented. While the appellant’s application for a patent was pending in the United States Patent Office, the appellee Mills filed an application entitled “Welding Composition,” which the appellees assert contains a disclosure of the alleged invention of the appellant’s patent.

The Patent Office, in allowing the appellant’s patent, did not declare an interference to exist between the appellant’s application and that of Mills, but granted the appellant’s patent while the application of Mills was still pending. In its brief, the appellant somewhat reluctantly admits that this was an “error” on the part of the Examiner, but adds that such “error” was almost “justified” by the “confused” “verbiage” of the application filed by Mills. After the issuance of the appellant’s patent, the appellee Mills caused an interference to be instituted between his application and the appellant’s patent.

At the time the oral argument was had before this court, the Patent Office had not yet held a final hearing on this question of interference. It seems tolerably clear, therefore, that the appellant’s patent was issued inadvertently. Such being the case, a different rule of law applies as to the presumption of validity. In Wilson & Willard Mfg. Co. v. Bole, supra, the late Judge Rudkin qualified the general statement above quoted in the following language: “But this rule 'is founded in reason.

It presupposes an adjudication by the Patent Office of every fact essential to the validity’ of the patent, and one who attacks that adjudication in a collateral proceeding must establish his claim by clear and satisfactory proof, or, as is often said, by proof beyond a reasonable doubt. But where it appears that there has been no such adjudication by the Patent Office, as whore the patent has been issued through inadvertence or mistake, the reason upon which the rule is founded ceases, and the rule ceases with it.

Section 4904 of the Revised Statutes (Comp.

St. 1913, § 9449 [35USCA § 52]) provides that: “ ‘Whenever an application is made for a patent which, in the opinion of the Commissioner, would interfere with any pending application, or with any unexpired patent, he shall give notice thereof to the applicants, or applicant and patentee, as the ease may be, and shall direct the primary examiner to proceed to determine the question of priority of invention.’ “Instead of declaring an interference between the Bole and Wilson applications to determine the question of priority of invention, the Patent Office inadvertently issued the Bole patent.

It would be illogical in the extreme to now hold that such a patent easts upon Wilson the burden of proving his ease beyond a reasonable doubt, whereas, if the same inadvertence had resulted in the issuance of a patent to Wilson, the like burden would fall upon Bole. On the contrary, it has been repeatedly held by the Patent Office and by the Supreme Court of the District of Columbia, in interference eases, that a patent issued through inadvertence or mistake casts no such burden upon the party claiming against it. 30 C’ye. 894, and eases cited.

The equity and justice of this rule is apparent and should control here. The fact that the trial court decreed in favor of the appellees on conflicting testimony is entitled to consideration; but if this court is convinced that the decree is erroneous, after giving due weight and consideration to the superior advantages possessed by the trial court, a reversal must follow.” Another leading case on the subject is American Soda Fountain Co. v. Sample (C. C. A. 3) 130 F. 145, 149, 150, certiorari denied 195 U. S. 634, 25 S. Ct. 791, 49 L.

Ed. 354, in which Judge Gray used the following language: “We do not agree with the contention, that the fact that the file wrapper discloses the patent to have been granted as first applied for, without any references, adds any force to the presumption of novelty arising from the grant. On the contrary, we think the force of that presumption is much diminished, if not destroyed, by the lack of any reference by the Examiner to, or consideration of, the ‘Clark’ patents. It does not seem likely that an expert examiner would pass them by, without notice or consideration, if they had been called to his attention.

We feel compelled, therefore, to the conclusion, that the first and fifth claims of the natcnt in suit are invalid for want of patentable novelty.” See, also, Westinghouse E. & Mfg. Co. v. Toledo, P. C. & L. R. Co. (C. C. A. 6) 172 F. 371, 392, 393; Elliott & Co. v. Youngstown Car Mfg. Co. (C. C. A. 3) 181 F. 345; 349; International F. Check Book Co. v. Young & Solden Co. (C. C. A. 4) 284 F. 831, 832; Goodbody v. Firestone Steel Products Co. (C. C. A. 6) 23 F.(2d) 625, 626; Standard Computing Scale Co. v. Lincoln Scale Corporation (D. C.) 57 F.(2d) 334, 336, 337; 48 C. J. 355, § 574.

We turn now to another preliminary question in this case; namely, that of estoppel. While admitting that, if the appellee Mills had merely filed an application on the same invention, that fact alone would not estop him and his assignee from attacking the patentability of the appellant’s invention, the appellant contends that “Mills did not stop with this,” and continues: “After the patent was issued to plaintiff [appellant], he [Mills] copied its claim and demanded an iivterference with it.

This amounted to an election on Mills’ part, for when he received notice of the plaintiff’s patent he had the option at that time to decide whether he would take the position that tlie supposed invention was not patentable or the position that it was patentable and he, Mills, was the first inventor of it.” (Italics are the appellant’s.) We cannot agree with this contention.

In the first place, inconsistent defenses are specifically permitted by Equity Rule 30 (28 USCA § 723), which provides in part as follows:' “The answer may state as many defenses, in the alternative, regardless of consistency, as the defendant deems essential to his defense.” See, also; Walker on Patents (6th Ed.) vol. 1, § 626, p. 720.

Nor does the demand for an interference in the Patent Office modify the general rule, correctly stated by the appellant; namely, that the filing of an application on the same invention does not estop the second applicant from attacking the patentability of the first applicant’s alleged invention. Numerous are the eases that sustain this view. For instance, in the leading ease of Holliday v. Piekhardt (C.

C.) 29 F. 853, 859, 860, the court said: “It has been urged that the defendants are estopped from contesting the validity of the claims of the patent in consequence of their action in the patent-office, and cannot recede from the position they then took, that the subject-matter was patentable, and that they were entitled to a patent because of priority of invention by Caro.

If the plaintiffs had been misled, or induced to take action or incur expense, in consequence of ^presentations or conduct on the part of the defendants which authorized them to suppose that they would obtain a valid patent if they succeeded upon the issue of priority, the doctrine of estoppel might be invoked. But the ease would be an exceptional one where a party who has prevailed upon one issue or defense in a litigation is es-topped from setting up a different defense in a subsequent suit brought by his adversary.

Such a ease might exist where the defense in the second suit is so inconsistent with that asserted in the first that both could not be true, or where the defense in the first suit was of a character to induce the plaintiff to change his ground of action, and bring a second suit. An interesting example of the latter class is found in the ease of Philadelphia, W. & B. R. Co. v. Howard, 13 How. 307 [14 L.

Ed. 157], where the defendant, having defeated the plaintiff in a prior action by asserting and maintaining that a paper in its possession was sealed with the corporate seal of the defendant, was not permitted, in a second action brought against it by the plaintiff, to defeat the action by proof that the seal was not affixed by the authority of the corporation.

But it is not true, as a general proposition, that a party, by putting forward one defense in a litigation, is precluded from asserting another against his adversary in a subsequent suit between them; nor can the general proposition be maintained that a contest in the patent-office upon the question of priority of invention will forever foreclose the defeated applicant for a patent from assailing the validity of the patent upon other grounds. In the present case there is no foundation for an estoppel, because both parties were fully aware of the prior state of the art before the interference was declared.

The plaintiffs could not, therefore, have been misled or prejudiced by the conduct of the defendants in attempting to defeat their application for a patent upon the ground that Caro was a prior inventor.” A similar view was expressed by Judge Lacombe in National Mach. Co. v. Wheeler & Wilson Mfg. Co. (C. C. A. 2) 79 F. 432, 441, 442, certiorari denied 160 U. S. 722, 17 S. Ct. 997, 41 L.

Ed. 1188: “It is contended by complainant that not only is the question of priority res adjudicata between the parties to the suit, but that defendant is also preeluded from contesting the validity and scope of claims 21 and 22 of the patent in suit, and infringement of the claims by defendant’s, machine.

The theory of this contention is that under the rules and practice of the patent office either party has a right to move to dissolve an interference on the ground that, in view of the state of the art, the issue framed therein could not be based upon his invention as described and claimed; that failure to move for such dissolution amounted to an acquiescence in the holding of the patent office that the inventions of the two parties as limited by the prior art there shown were identical; that they were patentable despite the prior art, and that either might properly be the basis for claims corresponding to the interference issues.

And therefore, that, although it now appears that the original Tebbetts & Doggett claims — now claims 21 and 22 — correctly cover defendant’s machine, but do not cover complainant’s unless an unexpressed element is read into them, defendant cannot' now avail of that fact to limit these two claims of the patent. We are referred to no authority in support of this contention.

It has not been the tendency of the decisions either of the supreme court or of the circuit courts or courts of appeal to extend the effect of interference decisions as final adjudications, and we concur with the circuit court in the conclusion that, 'while the decision in interference may be res adjudicata as to priority, it does not preclude defendant from raising other questions not in issue in said proceedings.’ ” See, also, Elliott & Co. v. Youngstown Car Mfg. Co., supra, at page 349 of 181 F., and 48 C. J. 181 and 191, §§ 262 and 284.

We have carefully examined the decisions referred to by the appellant in support of its argument on the question of estoppel. McCarty v. Lehigh Valley Railroad Co., 160 U. S. 110, 120,16 S. Ct. 240, 40 L. Ed. 358, and Steel Protected Concrete Co. v. Central I. & C. Co. (C. C.) 155 F. 279, 285; refused to sustain the claim of estoppel.

In the Steel Protected Case, Judge Saunders merely stated that he would hold in favor of estoppel “if there were authority for thus deciding this controversy as between plaintiffs and defendants;” but he actually ruled against the plaintiffs, although the alleged estoppel "would have lain against the defendants. Furthermore, the appellant admits that Judge Saunders’ remarks were dicta.

In the McCarty Case, the court was considering the effect of an award by the Patent Office on an interference; while in the instant ease, so far as the record shows, there was no decision on the interference proceedings in the Patent Office, hence no claim of estoppel arising from the result of those proceedings could here be put forth. There is certainly no authority for the contention that a mere interference proceeding works an estoppel. Uihlein v. General Electric Co. (C. C. A. 7) 47 F.(2d) 997, upon which the appellant relies “very strongly,” and Rice-Stix Dry Goods Co. v. J. A. Seriven Co. (C. C.

A. 8) 165 F. 639, likewise are distinguishable on the facts. We advert next to the question of the weight to be given by an appellate court to the findings of fact of a master in chancery, especially when such findings are approved and adopted by the District Court. In the instant case, it was stipulated that the master's report might constitute the findings of fact and conclusions of law in the cause. In the first place, the matter of invention is one of fact. In Thomson Spot Welder Co. v. Ford Motor Co., 265 U. S. 445, 446, 447, 44 S. Ct. 533, 534, 68 L. Ed. 1098, Mr.

Justice Sanford said: “The question whether an improvement requires mere mechanical skill or the exercise of the faculty of invention, is one of fact; and 'in an action at law for infringement is to be left to the determination of the jury. [Many cases cited.]” This court has applied the same doctrine in an equity case, Schumacher v. Buttonlath Mfg. Co. (C. C. A.) 292 F. 522, 533: “The decree adjudged claims 1 to 7 as being devoid of invention. This is also a finding of fact.” See, also, Ohmer Fare Register Co. v. Ohmer (C. C. A. 6) 238 F. 182, 187; H. D. Smith & Co. v. Peck, Stow & Wilcox Co. (C. C.

A. 2) 262 F. 415, 417; Kurtz v. Belle Hat Lining Co. (C. C. A. 2) 280 F. 277, 279; New Jersey Zinc Co. v. American Zinc, Lead & Smelt. Co. (C. C. A. 1) 284 F. 305, 308; O. K. Jelks & Son v. Tom Huston Peanut Co. (C. O. A. 5) 52 F.(2d) 4, 8. Since, therefore, the question of invention is one of fact, it next becomes necessary to' inquire into the degree of weight to be given to the master’s findings of fact, concurred in and approved by the District Court, in a general reference made as above set forth.

The books are well agreed upon the proposition that, under such circumstances, the master’s findings are to be given great weight. In Smith v. Hovland, 11 F.(2d) 9, 13, certiorari denied Smith v. United States, 371 U. S. 686, 46 S. Ct. 638, 70 L. Ed. 1150, this court said: “The findings of fact, having been approved by the District Court after a review of the evidence, are to be taken as presumptively correct, and unless obvious error has intervened in applying some principle of law or some important mistake has occurred in weighing the evidence, the decree will not be reversed. Furrer v. Ferris, 12 S.

Ct. 821, 145 U. S. 132, 36 L. Ed. 649; Road Imp. Dist. v. Wilkerson (C. C. A.) 5 F.(2d) 416.” See, also, on the general question of the weight to be accorded to the findings of fact of the trial court, McCullogh v. Penn. Mut. Life Ins. Co. (C. C. A. 9) 62 F.(2d) 831, and Markell & Co. v. Mutual Ben. Life Ins. Co. (C. C. A. 9) 63 F.(2d) 193, 194, 195.

Having cleared away the threshold difficulties of this case, we now advance to the main question; namely, was there manifest or “obvious” error in the master’s finding, adopted by the District Court, to the effect that the appellant’s welding tube was so lacking in invention as to be unpatentable? In considering this determinative element of the case, we must bear in mind that, where a patent is issued through “inadvertence” or “mistake,” the presumption of its validity is “much diminished, if not destroyed.” As was well said by Mr. Justice Brown, in Potts v. Creager, 155 U. S. 597, 606, 15 S. Ct. 194, 198, 39 L.

Ed. 275, the question “as to what invention really is” “has taxed the ingenuity of courts ever since the passage of the patent acts.” Nevertheless, the books contain certain guideposts that will aid us in our inquiry. As an attack on the inventive character of the device of the patent, the appellees introduced evidence by three witnesses that a so-called “hot rod” method had been developed by William B. De Long, a welder, during the latter months of 1926. This hot rod method consisted of taking pieces of tungsten carbide and laying these in a convenient receptacle.

A mild steel welding wire or welding rod was then heated at its end vrith an acetylene torch, and the molten end of the mild steel wire or rod was touched to one or more particles of tungsten carbide. The tungsten carbide would cling to the molten end of the rod, and the particle would then be transferred to the bit that was to be faced. The end of the mild steel wire was then melted off so that the tungsten carbide and mild steel would be deposited upon the bit.

The appellant admits that “there is a direct conflict in the evidence taken and in the evidence offered and refused by the Master as to who originated the ‘hot rod’ method.” The appellant further states that the master refused the testimony of Shelley M. Stoody to the effect that this method originated with the patentees within two years of the filing of their application, on the theory that, inasmuch as both the appellant and the appellees admitted that the hot rod method originated before the welding rod of the patent in suit, it was immaterial who originated it.

The appellant contends that this was error on the master’s part, “for if the ‘hot rod’ method was originated by the patentees, il. having been originated less than two years prior to the filing date of the plaintiff’s application, which is January 30, 1928, there was no actual! or constructive abandonment of this method to the public.” This asserted error on the master’s

part is made the basis of two assignments, as follows: “18. The Special Master erred in striking out "Witness Whaley’s testimony that the hot rod method originated with the patentees. “19.

The Special Master erred in refusing to accept Witness Shelley Stoody’s testimony that the hot rod method was originated with the patentees.” Assuming, but not deciding, that these two assignments of error may be based upon the ruling of the master rather than upon the ruling of the court in passing upon the exceptions to the master’s report, they do not conform to the requirements of rule 11 of this court, which provides in part as follows: “When the error alleged is to the admission or to the rejection of evidence, the assignment of errors shall quote the full substance of the evidence admitted or rejected.” Secondly, the exceptions to the master’s report do not point out specifically the alleged error of the master in striking out the testimony of Whaley and in refusing to admit the testimony of Stoody.

Assignment of error No. 19, dealing with the refusal to accept Stoody’s testimony, disregards Equity Rule 46 (28 USCA § 723), which provides in part as follows: “In all trials in equity the testimony of witnesses shall be taken orally in open court, except as otherwise provided by statute or these rules. The court shall pass upon the admissibility of all evidence offered as in actions at law.

When evidence is offered and excluded, and the party against whom the ruling is made excepts thereto at the time, the court shall take and report so much thereof, or make such a statement respecting it, as will clearly show the character of the evidence, the form in which it was offered, the-objection made, the ruling, and the exception.

If the appellate court shall be of opinion that the evidence should have been admitted, it shall not reverse the decree unless it be clearly of opinion that material prejudice will result from an affirmance, in which event it shall direct such further steps as justice may require.” We do not believe that the informal colloquy between the master and counsel in connection with a question asked of Stoody regarding De Long’s knowledge of how to pick up the particles of borium (the appellant’s tungsten carbide) with the rod amounts to conformance with the above Equity Rule.

Finally, in view of the state of the record and the amplitude of the evidence as to the lack of invention, we do not believe that the exclusion of the testimony covered by assignments 18 and 19, if erroneous, was prejudicial. The appellant, however, does not base its case solely or even chiefly upon the question of who originated the hot rod method.

In their brief, counsel say: “Even if we should concede for the purposes of argument that DeLong originated the ‘hot rod’ method, which is squarely in the face of Whaley’s testimony and Shelley Stoody’s offered testimony to the contrary, there must be a finding of invention under the law. Here we have a new combination of old elements.” The appellant lays great stress upon “the rule that a new combination of known or old elements which produces a new result or an old result in a better and more efficient manner” is evidence of invention.

We have no quarrel with this rule, for it seems well entrenched in the law of patents. Thus, in the leading ease of Webster Loom Co. v. Higgins, 105 U. S. 580, 591, 26 L. Ed. 1177, the court said: “It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention.” See, also, Seymour v. Osborne, 11 Wall. (78 U. S.) 516, 542, 20 L. Ed. 33; Gould v. Rees, 15 Wall. (82 U. S.) 187, 189, 21 L. Ed. 39; Hailes v. Van Wormer, 20 Wall. (87 U. S.) 353, 368, 22 L.

Ed. 241; Pickering v. McCullough, 104 U. S. 310, 317, 26 L. Ed. 749; Cantrell v. Wallick, 117 U. S. 689, 694, 6 S. Ct. 970, 29 L. Ed. 1017; The Barbed Wire Patent, 143 U. S. 275, 283, 284, 12 S. Ct. 443, 450, 36 L. Ed. 154; Leeds & Catlin Co. v. Victor Talking Mach. Co., 213 U. S. 301, 318, 29 S. Ct. 495, 53 L. Ed. 805; Diamond Rubber Co. v. Consol. Rubber Tire Co., 220 U. S. 428, 442, 443, 31 S. Ct. 444, 55 L. Ed. 527; Bliss v. Spangler (C. C. A. 9) 217 F. 394, 397; Bankers’ Utilities Co. v. Pacific Nat. Bank, supra, at page 18 of 18 F.(2d). But there are important qualifications to this rule.

First, it is not an invariable one. Webster Loom Co. v. Higgins, supra, and Knapp v. Morss, 150 U. S. 221, 227, 14 S. Ct. 81, 37 L. Ed. 1059. Again, such a new combination of old elements is not irrebuttable proof of invention, but merely, as we have seen, “evidence” of it. Webster Loom Co. v. Higgins, supra; Knapp v. Morss, supra; National Tube Co. v. Aiken (C. C. A. 6) 163 F. 254, 261. In the third place, in the language ,of Mr. Justice Strong, in Hailes v.

Van Wormer, supra, “the results must be a product of the combination, and not a mere aggregate of several results, each the complete product of one of the combined elements.” In other words, as was stated by the late Judge Gilbert of this court, in Pelton Water Wheel Co. v. Doble, 190 F. 760, 766: “In order to be patentable, a combination of elements must in their co-relation produce a different force, or effect, or result, from the sum of that which is produced by their separate parts.” See, also, Pickering v. McCullough, supra; Knapp v. Morss, supra.

In the fourth place, and most important of all, “an instrument or manufacture which is the result of mechanical skill merely is not patentable. Mechanical skill is one thing; invention is a different thing. Perfection of workmanship, however much it may * * * diminish expense, is not patentable. The distinction between mechanical skill, with its conveniences and advantages and inventive genius, is recognized in all the cases.” Reckendorfer v. Faber, 92 U. S. 347, 356, 357, 23 L. Ed. 719. See, also, National Tube Co. v. Aiken, supra, at pages 260, 261 of 163 F.; H. J. Heinz Co. v. Cohn, supra; Keszthelyi v.

Doheny Stone Drill Co. (C. C. A. 9) 59 F. (2d) 3, 7, 8. We turn now to apply the foregoing principles to the instant case. On the subject of the prior art and of the lack of invention, the master found as follows: “Summing up the prior art it is found that at the time of the appearance of the welding rod of the patent * * * “

(1) It was common practice to combine in rod form various steel substances intended for deposit in a weld and to use a steel tube filled with alloying substances for the purpose. “

(2) It was known that tungsten carbide could be used advantageously in hard surfacing cutting tools. “

(3) It was known that tungsten carbide was not materially affected by a temperature of the degree of the acetylene torch and that it formed a bond with mild steel or other matrix metals. * * * “It is true that the use of the tube of the patent results in a more facile and economical application of the material in a weld.

However, in view of the state of the art the step taken did not involve the necessary element of inventive thought, but was an improvement logically coming from workers in the art, who applied their skill and knowledge to a given problem.” There is ample evidence in the record to support these findings. Indeed, a goodly portion of the appellant’s brief is devoted to a discussion of conflict or significance of evidence, rather than the absence of it.

Thus, the appellant argues that “Mills’ attempt now to assert that his intention with the Oxite rod was to change all the tungsten into' tungsten carbide and to produce a heterogeneous deposit, cannot be reconciled with his actions”; that “certainly Mr. Stone would not have told W. F. Stoody, who he knew was a welding man engaged in the welding business, that Stone’s men were drilling holes and calking if they had prior to Feb. 19,1927, been welding the Thoran [tungsten carbide] on”; that “as against this uniform testimony of plaintiff’s witnesses W. F. Stoody, Shelley M.

Stoody, Whaley and defendants’ own witness-' es Norman W. Cole, whom the defendants had persuaded to testify against his own patent, the defendants produce. Stone, Mayer, DeLong and Edmond.

The first three, Stone, Mayer, and DeLong, are highly biased,” etc.; that “DeLong then proceeds to let his imagination run riot and tell how Stone brought out some pieces of Thoran,” etc.; that a certain portion of Edmond’s testimony “is not corroborated by anybody,” and is “peculiar”; that the defendants’ “own witness Cole contradicts Mayer and DeLong,” etc., and that “there is no way of either corroborating or disproving Edlund’s knowledge or lack of knowledge that Borium had been placed in . mild steel tubes prior to that time by the patentees,” etc.

Finally, the appellant even devotes two pages of its brief to a

summary, in parallel columns, of the testimony of the appellant’s and the appellees’ witnesses on the question of whether or not the welding of tungsten carbide by playing the torch directly upon it while melting adjacent mild steel was discovered by the patentees. This somewhat surprising

summary is preceded by the frank statement, “Let us weigh the evidence.” We have referred at some length to the foregoing passages in the appellant’s brief because we regard them as significant in two respects: First, they indicate a misconception of this court’s duty with regard to considering the weight of the evidence adduced before the master; and, second, because the very discussion in the appellant’s brief discloses that there was some substantial evidence in support of the master’s findings.

In fact, as we have seen, the appellant frankly extends to us the invitation, “Let us weigh the evidence.” But “weigh the evidence” is precisely what an appellate court, in the present situation, cannot do. We ha ve considered the other assignments, but find that they disclose no reversible error. Indeed, most if not all, of the points involved in such assignments have already been discussed in this opinion.

Therefore, bearing in mind the weight to be accorded to the findings of fact of the master, concurred in by the District Court, on a general reference, and keeping before us the rule that invention is a matter of fact, we hold that the record discloses substantial evidence to support the judgment of the court below. That judgment is accordingly affirmed. Judgment affirmed.