American Stainless Steel Co. v. Rustless Iron Corp., 2 F. Supp. 742 (D. Md. 1933)

*743 Marshall A. Christy, of Pittsburgh, Pa., Charles Neave and Maxwell Barus, both of New York City, and Semmes, Bowen & Semmes, of Baltimore, Md., for plaintiffs.

C. P. Byrnes, G. E. Stebbins, and Walter J. Blenko, all of Pittsburgh, Pa., and J. Kemp Bartlett and Edgar Allan Poe, Jr., both of Baltimore, Md., for defendant.

WILLIAM C. COLEMAN, District Judge.

This suit is one for infringement of two patents, relating to stainless steel: Clement, No. 1,365,091, application filed March 12, 1917, issued January 11, 1921, which is for a product, that is, an alloy; and Hamilton and Evans, No. 1,432,289, application filed May 23, 1922, issued October 17, 1922, which is for a process, that is, a method of reducing metals and making alloys.

The principal plaintiff, the American Stainless Steel Company, is not a manufacturer, but a patent-holding company, which grants licenses to manufacturers, under its various patents, to make stainless steel. The other plaintiff, the Electro Metallurgical Company, a limited manufacturer, was found by the court at the inception of the trial, as a result of a motion by defendant to dismiss the suit because of a deficiency of necessary parties plaintiff, to be a necessary party plaintiff, since as part of the transaction by which it had assigned to the American Stainless Steel Company the two patents here in suit, it had retained for itself certain interests in their subject-matter. The defendant company is a manufacturer, operating a plant in Baltimore since 1926. With respect to both patents, the defendant resists the suit with numerous defenses, all of which may be included under the usual (1) denial of infringement, and (2) invalidity of the patents. The burden of proving infringement is upon the plaintiffs. The burden of proving invalidity is upon the defendant. We will consider the two patents separately, and will first take up the Clement patent.

Claims 10 and 11 only of the Clement patent are here involved. This invention, as stated in the patent, relates "to alloys and has for its object the provision of a composition of this character which shall be susceptible of easy working by commercial cutting, boring, turning and shaping tools and devices; which shall be highly resistant of chemical action, corrosion or oxidation, as by the action of acids, alkalis or oxidizing atmospheres," and "consists of iron and chromium, the chromium content being equal to at least 10% of the whole and the carbon content being not greater than .2 of 1% of the whole and preferably not over about one-tenth of 1%. * * *" His "improved alloy is not a chrome steel, the criterion of chrome steel being the possession of a high carbon content which produces a substance of such hardness and brittleness as absolutely to prevent any working or machining operations, this hardness being retained even to a red heat. My researches have shown that if the total carbon content be less than .1 or .2%, this hardness is entirely absent and the resulting metal possesses exactly the qualities of chemical resistivity combined with mechanical working ability which it is the object of my invention to secure." Specifications, p. 1, lines 10-15, 31-37, 45-59.

Claim 10, which is the more specific of the two claims here in suit, is as follows: "An alloy containing iron and chromium wherein the chromium constitutes not less than 10% and the carbon less than .2 of 1% of the alloy and substantially free from oxids." The only other claim in suit, claim 11, specifies the composition of the alloy and identifies its characteristics, namely, it is an "alloy of iron with upward of 10% of chromium which is ductile and malleable when cold and susceptible of machining and substantially free from oxids." By the disclaimer filed in the case, both of these claims are further limited as to the carbon content to not more than about one-tenth of 1 per cent.

The specifications describe the process of melting the iron and the chrome ore which, summarized, is as follows, the process under this patent, however, not being at issue: The iron is thrown in a mass into an electric furnace, into which protrude electrodes, through which the electric current forms an arc through the metal, thereby melting down all of the metal until it becomes fluid, and the *744 slag remains floating on the top of this lower, heavier metal.

It will be seen that the novelty claimed by Clement is that he found he could make iron or steel stainless, even if not hardened, if he kept the amount of carbon very low and lowered the oxids. He was not the creator, however, of stainless steel. Harry Brearley, an Englishman, and Elwood Haynes, an American, were the real pioneers, although for perhaps the previous half century the experiments and discoveries by others in the steel industry in England had been strongly prophetic of what these men actually did. To the former was issued a patent for cutlery as early as 1916, and to the latter a patent for a wrought metal article as early as 1919. Both related to the composition and production of acid resisting or stainless steel for the manufacture of cultery or other articles where nonstaining properties were desired. Haynes covered alloys containing carbon of from .1 per cent. to 1 per cent. and chromium of from 8 per cent. to 60 per cent. Brearley covered alloys containing carbon of less than .7 per cent. and between 9 per cent. and 16 per cent. of chromium. In an infringement suit, American Stainless Steel v. Ludlum Steel Co., decided in 1923 by the Circuit Court of Appeals for the Second Circuit, 290 F. 103, both of these patents were declared valid. They were also held to have been infringed by a high chromium-silicon alloy, with an average of carbon, .43 per cent. The practical scope of these two patents may best be understood from the following description taken from the court's statement of facts in this suit (page 104 of 290 F.): "Haynes called his application one for a `wrought metal article,' and Brearley his for `cutlery.' But the object of both patentees is the same, and may be shortly described as a desire to produce what has for some years been increasingly well known under the trade-name `stainless steel.' Although Brearley's patent date is earlier, his date of application is later, and it may be summarily held that Haynes' is the generic and Brearley's the specific patent. Haynes declares that his invention relates to the production of wrought metal articles of manufacture `having polished surfaces of the general character which is termed noble, in that such surfaces are incorrodible.' Brearley states that his invention relates to the production of `cutlery or other hardened and polished articles of manufacture where nonstaining properties are desired.'" The court concluded its opinion with the statement: "Whether there may not be in Haynes an attempt to cover too much ground is a question that may be left until occasion for decision arises." We assume that this was intended to refer primarily, if not exclusively, to the wide range of chromium.

In the decision on exceptions made by both parties to the master's report on the question of accounting in this same case (American Stainless Steel Co. v. Ludlum Steel Co. (D. C.) 16 F.(2d) 823), the court affirmed the master's inclusion, in the accounting, of all steels which were within the field of percentages of carbon and chromium common to both patents, and which were sold to manufacturers of cutlery, with one exception found not to be within the Brearley patent, and which we need not dwell upon here.

The Brearley and Haynes patents were the foundation of the business of the American Stainless Steel Company, whose licenses under these two patents include a large number of the leading steel manufacturers in the country.

Then came Clement. He was an employee of the Cleveland Brass Manufacturing Company at the time he filed application for his patent in suit. The Brearley United States patent had already been issued, but the Haynes application was still pending. Clement died in 1919, almost two years before his patent issued. He claimed to have made an advance in the pioneer work of Haynes and Brearley, that is to say they relied, as we have seen, upon hardening to make their steels stainless, while he claimed he had discovered that he could do so even if the steels were not hardened, if he kept the carbon content as well as oxids very low; that is, he would still continue to have a soft steel, workable when cold. New fields were opening for stainless steel, until now its use has become very broad in both exterior and interior building construction work, in certain types of valves, in the bright metal parts of automobiles, and for a variety of other purposes.

We will first consider the defense of invalidity of the Clement patent, in support of which defendant asserts it is invalid because of (1) the disclaimer; (2) disclosures in prior patents and especially in Haynes' patent, No. 1,299,404, heretofore referred to; (3) prior public use; (4) disclosure in prior publications; (5) inoperativeness of the process in Clement to produce the claimed result; and (6) insufficient disclosure in Clement for production of a ductile and malleable metal. We will take up these defenses in the order named.

On February 9, 1932, the Steel Company filed in the Patent Office a disclaimer in due form "of so much of the subject matter" of the two claims here in controversy "as may cover or include a product otherwise conforming to the terms thereof and having a content of carbon over about one-tenth of one per cent." When the Electro Company was made a party plaintiff, it filed a similar disclaimer, all of which was done pursuant to the provisions of sections 4917 and 4922 of the Revised Statutes (35 USCA §§ 65, 71).

The reasons stressed by defendant in support of its argument that the disclaimer invalidates the claims are two: First, that the filing of the disclaimer was unreasonably delayed; and, second, that the disclaimer fails to distinguish, with sufficient definiteness, between what is retained and what is disclaimed.

As to the first ground, namely, unreasonable delay, the burden of proof is on the plaintiffs to show that there was no such delay in disclaiming. The reason given by the plaintiffs for filing the disclaimer was that certain of the claims were found to have been inadvertently made "too broad in including that of which the said Alvah W. Clement was not the original and first inventor." Or more specifically, as was stated, the disclaimer was filed as a matter of policy, there being a divergence of opinion between counsel representing the plaintiffs, whether, with the claims as originally asserted, there was debatable ground.

Plaintiffs claim that a disclaimer is seasonable if made at any time before a judicial decree of invalidity. They deny that the disclaimer is a confession that the original Clement claims were void, either as anticipated by Haynes or otherwise; that where the excess is not apparent at once upon inspection, the allowance of a claim by the Patent Office raises such a presumption in its favor that the patentee may rely on its validity until a court of competent jurisdiction decides that it is broader than his real invention. On the other hand, defendant claims that disclaimer must be made as soon as the patentee becomes aware that he has claimed more than he has invented or described; that here the element of good faith is lacking; and that therefore those decisions which support the right of a patentee to postpone disclaimer until after a court decision have no applicability.

But we do not find that the proof supports defendant's contention. On the contrary, we believe plaintiffs have sustained the burden of showing that under the circumstances the disclaimer was not too late. See O'Reilly v. Morse, 15 How. 62, 14 L. Ed. 601; Seymour v. McCormick, 19 How. 96, 15 L. Ed. 557; Metropolitan Device Corp. v. Cleveland Elec. Illuminating Co. (C. C. A.) 36 F.(2d) 477. We do not find anything in the recent case of Ensten v. Simon, Ascher & Co. Inc., 282 U.S. 445, 51 S. Ct. 207, 75 L. Ed. 453, which supports defendant's contention under the existing facts. Indeed, there the doctrine of the earlier cases relied upon by plaintiffs is expressly affirmed in the following language, page 453 of 282 U. S., 51 S. Ct. 207, 209: "`The question of unreasonable delay is a question for the court, upon the facts as found either by its own investigation or the verdict of a jury. Delay begins whenever the patentee becomes aware that he has claimed more than he has invented or described. In cases where the excess is not apparent at once upon the inspection of the patent by the patentee, the allowance of his claim by the patent office raises such a presumption in its favor that he may rely on its validity until a court of competent jurisdiction decides that it is broader than his real invention.' Robinson on Patents (1890) vol. II, page 284."

Coming next to the second ground urged by defendant for the invalidity of the disclaimer, namely, that it fails definitely to distinguish between what is retained in the patent and what is disclaimed, we also feel that as to this point the plaintiffs have adequately sustained the burden of proof.

Plaintiffs maintain, and we think correctly, that in patents relating to the metallurgy of iron and steel, where composition limits are defined as well as where qualitative properties are characteristics, there is necessarily present, at about the limit points, what may be termed a twilight zone; that is to say, that a mathematical line cannot be drawn which will, for all purposes, accurately define either the upper or lower composition limit. They claim this is particularly true with respect to carbon content, where relatively small differences may greatly affect the properties of the product, and where the personal equation of the analyst necessarily enters. Dr. Brophy, analyst of the Research Department of the General Electric Company, testified that the limit of correct carbon analysis should be two one-hundredths of one per cent. either way. Plaintiffs further point out that the use of the adjective "about" in the narrowing of its claim is the adoption of the very term used in the Clement patent, where it is stated, line 31, page 1, "essentially *746 my improved alloy consists of iron and chromium, the chromium content being equal to at least ten per cent of the whole and the carbon content being not greater than two-tenths of one per cent of the whole and preferably, not over about one-tenth of one per cent." Thus the patent specification described a preferred content, but permitted an alternate content, and it is this latter which the disclaimer renounced, and limited the patent to the preferred content. Where such alternatives exist the patentee may disclaim one, and retain the other, provided the claim as so modified is supported by adequate disclosure in the specifications, and does not result in the substitution of a new claim. Hailes v. Albany Stove Co., 123 U.S. 582, 8 S. Ct. 262, 31 L. Ed. 284; Hurlbut v. Schillinger, 130 U.S. 456, 9 S. Ct. 584, 32 L. Ed. 1011; Electrical Accumulator Co. v. Julien Electric Co. (C. C.) 38 F. 117; Simplex Railway Appliance Co. v. Pressed Steel Car Co. (C. C. A.) 189 F. 70; Permutit Co. v. Harvey Laundry Co. (C. C. A.) 279 F. 713; Permutit Co. v. Wadham (C. C. A.) 13 F.(2d) 454; Sachs v. Hartford Electric Supply Co. (C. C. A.) 47 F.(2d) 743.

That there is some difference between steels of the specified chromium composition having carbon contents, or slightly below, and those having them slightly above, .10 per cent. is clearly supported by the weight of the evidence. Defendant's steels are of two classes, so-called Defirust and Special Defirust. Defirust has a chromium content of 12 to 14 per cent., the Special Defirust of 16 to 18 per cent. Dr. Waterhouse, one of plaintiff's experts, had two samples of each grade made, one with carbon below .10 per cent., the actual content being .07 per cent., and the other with a carbon content above .10 per cent., the actual carbon content being .13 per cent. All four of the steels were subjected to the usual qualitative tests, and to corrosion tests also, in three states: First, "as rolled," second, in the "hardened" condition; and, third, when tempered after hardening, to give maximum softness. His conclusions were that there is a considerable difference in the properties of these chrome steels, if the carbon is slightly below .10 per cent. and if it is slightly above .10 per cent., the higher carbon steel being very distinctly harder, higher in strength, considerably lower in ductility; in short, that the lower carbon steels were more malleable and ductile when cold.

The same results are shown by defendant's analyses of its own steels shipped to the Superior Steel Corporation during 1928-29; that is, that over 90 per cent. of these shipments contained .10 per cent. carbon or less. Similarly, in defendant's trade catalogue, the carbon content of both Defirust and Special Defirust is given as "under 10," and the United States Steel Corporation advertised its two rustless iron products as having a maximum carbon content of .10 per cent. Also, it appears that the maximum carbon content in the specification of the General Electric Company, for the steel which it obtains from the Alleghany Steel Company for its turbine bucket blades, is .12 per cent. As plaintiffs point out, these are not mere accidental contents.

It is true there is a good deal of evidence introduced by defendant which is contrary to the above testimony of their experts, Drs. Sauveur, Berry, and Grinnell, also some published government and commercial specifications. But plaintiffs are not required to prove that every one skilled in the art, or conversant with the trade which deals with the art, shall agree on the value or importance of this technical carbon differentiation. It is sufficient, in order to support the disclaimer and that is the only point we are now considering if the plaintiffs show that the differentiation is recognized in any substantial degree by the art and the trade, as a material one. This we believe plaintiffs have shown. They do not claim that there is a sharp break in the curve at a mathematical line of .10 per cent. carbon, but merely that there is a definite zone of demarcation at "about one tenth of one per cent." of carbon. The patentee, Clement, believed this to be true; numerous not all producers and users of the product have adhered to this belief. Necessarily, in weighing the evidence on a question of this sort, we find ourselves entering, to a large extent, the field of individual opinion, expert and otherwise. So we must give to the evidence as a whole a reasonable construction, after applying both the qualitative and quantitative analysis. Thus defendant may actually have found, as one or more of its witnesses testified, that the commercial limits, with respect to its own products, for rustless or stainless iron, are .05 per cent. to .15 per cent. carbon. But this does not compel us to discard or to subordinate other evidence from sources which we may feel to be equally, or more credible.

We now come to defendant's second ground of defense; that is, disclosure in prior patents, especially in Haynes.

*747 The Haynes patent issued April 1, 1919, subsequent to Clement's filing date (March 12, 1917), but the Haynes application was filed March 12, 1915, two years prior to Clement's application. Thus, although when Clement filed his application he had no knowledge of Haynes, nevertheless Haynes is a proper reference against Clement. See Alexander Milburn Co. v. Davis-Bournonville Co., 270 U.S. 390, 46 S. Ct. 324, 70 L. Ed. 651.

A presumption of novelty arises from the grant of a patent by the Patent Office. But this presumption may be at least weakened by the fact that pertinent references were not cited against it during its progress through the Patent Office. Haynes was not cited against Clement. See Cleveland Foundry Co. et al. v. Kaufmann Bros. (C. C.) 126 F. 658; Westinghouse Elec. & Mfg. Co. v. Toledo, P. C. & L. Railway Co. (C. C. A.) 172 F. 371; Elliott & Co. v. Youngstown Car Mfg. Co. (C. C. A.) 181 F. 345; American Soda Fountain Co. v. Sample (C. C. A.) 130 F. 145.

Let us now analyze the range of the Haynes patent, both with respect to carbon and chromium content, in relation to Clement.

The limits in Haynes are from .1 per cent. to 1 per cent. carbon, and from 8 per cent. to 60 per cent. chromium. Thus the Haynes carbon range, since its minimum includes .1 per cent., meets Clement. Haynes defines (page 1, line 29) his preferred carbon range as "between 0.1 and 0.5 per cent." As to chromium, the Clement range is "not less than 10%." Haynes specified (page 1, line 22) that his alloy should contain "not less than 8% chromium and, very advantageously, not less than 10% and not more than 50 or 60 per cent. the best proportion of chromium being between 15 and 25 per cent." Again Haynes says (page 3, line 94) that "I consider the alloys containing 20 to 25 per cent. chromium with 0.1 to 0.5 per cent. carbon as best adapted for the present purposes."

Turning next to the question of freedom from oxides, Haynes specifies that his alloy should be deoxidized (page 2, lines 91-102). A metal containing a large amount of oxides will not forge properly when hot. Haynes says (page 1, line 46) that his metal is capable of being easily worked and manufactured as by forging, hammering, or otherwise working. So it follows that Haynes metal must be "substantially free from oxids."

Next, with respect to ductility and malleability, Haynes (page 3, line 70) recites that his product may be readily drawn or rolled into rod, wire, sheets, etc. This quality necessarily implies at least some degree of workability, while cold softness, ductility, malleability.

Lastly, as to the rustless or stainless quality claimed for Clement, Haynes states that the polished surfaces of his articles are "incorrodible, lustrous and of permanent nature" (page 1, line 14); and he describes in more detail the extent of their susceptibility to various enumerated corrosive agents (page 2, lines 103-110; page 3, line 50).

In spite of the aforegoing disclosure by Haynes of the full composition of Clement, plaintiffs contend that there is a material differentiation which must be given legal recognition. They contend that Haynes teaches only a hard material; that he said he could go as low as .10 per cent. in carbon content, but no lower; that all of the Haynes claims are limited to a minimum carbon content of .10 per cent.; that while in composition only, there is an overlap of one tenth of a per cent. of carbon between Clement's original maximum of .20 per cent. and Haynes' minimum of .10 per cent., nevertheless, plaintiffs do not admit that what they disclaimed was old in Haynes, and was not new in Clement. Plaintiffs assert that whatever may be said as to the Haynes specifications of carbon content, Haynes neither had in mind nor described any product such as Clement's which is ductile and malleable when cold and susceptible of machining, but, on the contrary, an essentially hard tool steel product, such as a knife blade, and therefore one which is malleable and ductile only "at high, metal working temperatures" (page 1, line 101).

It is true that the Haynes specifications relate specifically to articles "of the nature of cutlery and edged tools" (page 1, lines 8, 36, 89, 100), and it is urged that while the Haynes alloy is malleable when heated and is capable of being readily forged and worked while hot, hardening on cooling (page 2, line 103), it can be forged, worked, or shaped only at high temperatures, that is, it can be cold-drawn into wire only if it is first annealed, whereas Clement's alloy can be drawn into exceedingly fine wire, without any preparatory annealing. But is this a fact? Can it really be so, if, as is beyond contradiction, from the letter of the specifications themselves, the permissible content under both patents, of both carbon and chromium, may coincide? Both bald logic and the weight of the credible, technical evidence would seem to compel a negative answer to this question.

*748 In the first place, it appears that the Alleghany Steel Company, a licensee of the plaintiff Steel Company, under the Haynes patent, has paid to the Steel Company royalties under this license on products having carbons as low as .04 per cent.; that the Alleghany Steel Company has never made any iron chromium alloys hardenable to knife hardness; and that what it did make was malleable, ductile, and machinable when cold. Furthermore, Dr. Waterhouse, when asked how many of the alloys would be knife steels, assuming that, by following the Haynes preferred range, you took a range of alloys with carbon .1 per cent. and the chromium range from 15 per cent. to 25 per cent., replies: "Only those towards the low percentage of chromium * * * I do not believe that in these chromium steels you could get a satisfactory knife blade with .10 per cent. carbon."

Again, it is not without some significance that the companion case (Equity No. 1589) filed in this court, but later dismissed, charged that the same Defirust material of defendant, which is here alleged to infringe Clement, infringed Haynes.

The Circuit Court of Appeals in the Ludlum Steel Company Case, supra, dealt fully, and correctly we think, with the question of temperature variations, and said, pages 107, 108, of 290 F.: "These [Brearley and Haynes] patents refer to the production or making of a certain kind of steel; they instruct with remarkable precision as to the component parts of the material, which is to be stainless. It seems to us clear that such a disclosure speaks directly to those whose business it is to make the steel, and not to the cutler who is to fashion it. * * * It follows that we are of opinion that these patents are to be interpreted in the light of what their disclosures would mean to one accustomed to the art of making steel, or to a metallurgist, and not to one who was concerned only with the forming thereof into implements of industry. * * * Haynes thought that he could use either electric or fire heat, while Brearley preferentially recommended an electric arc melting furnace. * * * It is admitted that both the patents plainly disclose a hard steel, particularly when the chromium is low, and it is also admitted that the process of hardening generally is not a matter of heating to a fixed point. While it has long been known that the critical point of any steel may be indicated by pyrometric instruments, well known in commerce, it is also proven that even among artisans, including cutlers, the practice of skilled men has been to determine the proper hardening or heating point empirically. We do not go into the exact number of heat degrees above the critical point of ordinary steel, to which steels like defendant's or Brearley's preferred product (minus the silicon) must be heated to insure stainlessness, because we do not think the exact number of degrees makes any difference, and are of opinion that such higher degree of heat is very far within the professional knowledge of the metallurgist or steel maker by whose competency these patents are to be judged. Indeed, we are of the opinion that the determination of such proper degree of heat is within even the range of a workman's competency under what has been called the `try it and see' rule. Cf. Bethlehem, etc., Co. v. Niles, etc., Co. (C. C.) 166 F. 880, affirmed 173 F. 1019, 98 C. C. A. 640; Pittsburgh, etc., Co. v. Seaman Co., 248 F. 705, 160 C. C. A. 605."

At .10 per cent., by the limits expressly provided in the specifications in both Haynes and Clement, the carbon contents meet. Plaintiffs, as we have seen, contend, in support of their claim of infringement, that a leeway, both up and down of .02 of 1 per cent. is impliedly indispensable in the nature of things both from the metallurgical or scientific aspect, and the personal equation aspect. Obviously this leeway should be allowed in a consideration of the patent's validity, just as much as it should be allowed when considering whether the patent has been infringed. So the plaintiffs are not consistent when they claim that, under Haynes, it is permissible to go as low as .10 per cent. in carbon content, but no lower, albeit it is not actually necessary to go lower to produce a parallel composition.

Thus, in conclusion, while, as we have pointed out for the purposes of disclaimer per se, it is sound to say that since, between a carbon content of "about .10 per cent." and a carbon content of .20 per cent., there exists a doubtful and uncertain field wherein the Clement invention, as originally defined by him, may or may not be capable of realization, this doubtful field should be eliminated, nevertheless it is quite another thing to say that such elimination successfully disposes of the distinctly separate question, namely, whether Haynes has in fact anticipated, and thereby invalidated, Clement, when, as is true, after giving full force and effect to the disclaimer, Haynes and Clement "meet" as to carbon content at .1 of 1 per cent., and when also the chromium range of Clement is well within that of Haynes.

These patents describe an alloy of iron and chromium claimed to be highly resistant to attack by ozone, either alone or in the presence of moisture. The chromium content is specified as being from 25 per cent. to 40 per cent., and the combination of the chromium and iron shall "be as pure as possible and especially if they be free, or as nearly free as possible from carbon." Thus, this iron chromium alloy, since it has a chromium content as high as 25 per cent., may be said to come within the contemplation of the Clement patent; but since the Clement patent requires the presence of carbon in an amount not over about .10 per cent., and since these other patents not only do not attempt to specify a minimum of carbon, but on the contrary provide that the combination shall be as free as possible from all carbon content, it is at least open to serious question whether these foreign patents can be said to teach the Clement alloy.

We turn now to the next argument advanced for the invalidity of the Clement patent, namely, its prior public use, which is alleged to have been of two kinds; that is to say, in alloys of the General Electric Company used (1) for lead-in wires for electric lamps, and (2) turbine blades.

It appears that the lamp wire development was the first, and arose directly on account of a commercial problem which confronted the General Electric Company. The purpose of a lead-in wire is to carry electric current into and out of the glass bulb of an incandescent lamp. The wire must go through the glass and must have a coefficient of expansion approximately the same as glass, and must be resistant to corrosion. Edison, the pioneer inventor, used platinum as a lead-in wire, which has the two essential qualifications, namely, coefficient of expansion equal to glass and resistance to oxidation. On account of the cost, however, of platinum, in 1910 the General Electric Company sought a substitute for it. Following laboratory experiments by Dr. Dantsigen, a member of the staff of the General Electric Company's Research Department at Schenectady, alloys were made with a chromium content varying from about 13 per cent. to about 29 per cent., and with a carbon content from .1 per cent. to .5 per cent., which were easily malleable at high temperature and could be machined and worked at ordinary temperature. Two of these showed a carbon content of .1 per cent. and ten showed a carbon range from .1 per cent. and .2 per cent., both inclusive. As a result of these experiments, in the latter part of 1912, a large number of Christmas tree lamps, that is, more than 100,000, were made up and sold, and approximately 50,000 feet of the wire were used in the lamp development. Thereafter, however, its use was discontinued because of the development of the so-called Dumet wire, which proved more satisfactory.

This lamp wire development led directly to the turbine blade development which may be summarized as follows: Dr. Whitney, head of the Research Laboratory of the General Electric Company, followed up the work of Dantsigen on the lead-in wire. The alloy was made in larger quantities, but the experiments at the factory where the alloys were tested as small turbine bucket material, were discontinued due to a variety of difficulties which were encountered, and in 1915 all of the buckets there made were scrapped. Meanwhile, however, corresponding experimental work was progressing at the factory at Schenectady, where the larger turbines were made. The chrome iron alloy was made into large buckets and put into a turbine in 1915, which went into operation the following year for the purpose of tests, and the turbine was operated, without change, until 1919 when it was opened for inspection. The blades were then found to be still free from corrosion, were repolished, and replaced, except some 35 out of 300 of the blades, which were found to have become seamed, and these were replaced with nickel steel blades. Thereupon the turbine was put back into operation, and continued in use for another three and one-half years, until 1923, when the second inspection was made. Again they were found to be in excellent condition. The turbine was closed up again, put back into service, and is still in use supplying power for the General Electric Company and the New York Power and Light Company.

The position taken by plaintiffs with respect to the use of the alloy for lead-in wires for electric lamps is that it was so remote from the uses to which the Clement product is put, that these prior lead-in wires, whatever they may have amounted to, cannot affect the validity of the Clement patent. As to the use of the alloy for turbine blades, plaintiffs do not deny that it has been proven that the General Electric Company did make a considerable quantity of steel having the composition and the qualities which defendant claims for it, and that some of this steel was used in one turbine. However, plaintiffs take *750 the position that this manufacture and use was experimental only; that the experiment turned out to be a failure; that it was in effect abandoned at the time when the turbine containing these blades was first opened in 1919; and that the experiment taught the art absolutely nothing. Plaintiffs contend that while the existence of the blades was well known to a considerable number of highly trained technical employees of the Research Department of the General Electric Company, none of them at any time turned the information to a practical purpose. By way of corroboration, plaintiffs stress the further fact that the General Electric Company later bought its alloys from the Alleghany Steel Company, and further tested that company's own low carbon high chromium steel product, which it ultimately adopted.

We are not impressed with these contentions of the plaintiffs. So far as the lead-in wire is concerned, the essential fact remains that the alloy was used in it, and used commercially. The fact that, later on, a better substitute was found, is not controlling. With respect to the use of the alloy in the turbine blades, while they were never sold, and while it may be conceded that the use was, and still is, merely experimental, this does not constitute the experiment an abandoned one. Commercial use is not essential. See Corona Co. v. Dovan Corp., 276 U.S. 358, 48 S. Ct. 380, 72 L. Ed. 610; Coffin v. Ogden, 18 Wall. 120, 21 L. Ed. 821; Brush v. Condit, 132 U.S. 39, 10 S. Ct. 1, 33 L. Ed. 251; Bliss Co. v. Southern Can Co. (C. C. A.) 265 F. 1018; Id. (D. C.) 251 F. 903. The use of the turbine began in 1916 and still continues. When the turbine was opened for inspection in 1919, after three years' service, the majority of the buckets were found to be in good condition, were repolished, put back, and are still in use.

Lastly, the fact that the General Electric Company later bought alloys from the Alleghany Steel Company is not to us convincing as an argument that what was done with respect to this turbine, commencing in 1916, was in effect merely an abandoned experiment. The Alleghany Steel Company's alloy is shown to be a chrome iron alloy, having a maximum of .12 per cent. carbon, and of chrome from 11.5 per cent. to 13.5 per cent., thus closely approximating the alloy in the turbine. The latter was known as Turbuck metal, the former as Ascoloy, and we believe from examination of the evidence that it may truthfully be said, as defendant contends, that the evolution of the General Electric Company's experiments proved that the use of Turbuck was a link in the chain of development and use of Ascoloy.

At this point it may be asked whether, upon the view just taken of the prior uses of the lead-in wire and the turbine blades with respect to Clement, these prior uses would not also have invalidated the Haynes patent, because the use of the lead-in wire at least, if not also of the turbine blades, appears to have antedated the Haynes application. Suffice it to say in reply that, presumably, if in the Ludlum Company Case, supra, Haynes had claimed the alloy as Clement did, and if one or both of these prior uses had been made a matter of evidence in that suit, the Haynes patent might have been held invalid. But on the present record, we do not question the validity of the Haynes patent, as determined by the Circuit Court of Appeals for the Second Circuit in the Ludlum Company Case, supra.

In further support of the argument of prior public use, defendant advances the analysis of some small quantities of the alloy which Marsh, a well-known inventor of alloys for electrical resistance wiring, furnished the defendant under his patent No. 1,057,784. In view of the fragmentary character of the testimony of Marsh, entirely by deposition, and especially the absence of any disclosure as to the exact carbon content, we consider that this part of the evidence is not entitled to very serious consideration. The same may be said regarding the small experimental in-got of the Firth-Sterling Steel Company, which appears to have been no more than an abandoned experiment.

In view of our conclusions respecting disclosure by prior patent, and prior public use, we deem it unnecessary to dwell at any length upon the three remaining arguments advanced by defendant in support of its claim of invalidity of the Clement patent, namely: (a) Disclosure in prior publications; (b) inoperativeness of the process to produce the claimed result; and (c) insufficient disclosure in Clement for production of a ductile and malleable metal.

With respect to disclosure in prior publications, reliance is placed primarily upon an article by Monnartz in the French technical journal "Metallurgie." Investigations made by this scientist appear to have been made with chromium-iron alloys from which the carbon was intentionally excluded. This fact alone, not to mention the fact that the investigations appear never to have advanced beyond a purely experimental stage, is sufficient to justify little emphasis being placed upon *751 this type of disclosure. Respecting the other printed publications referred to, suffice it to say that evidence regarding purely laboratory experiments therein given fails to show that the product of these various analyses did in fact embrace the same salient characteristics of Clement, especially as to noncorrosive quality.

Respecting the contention that the Clement process is inoperative to produce the claimed result, and that the disclosure in Clement is insufficient for the production of a ductile and malleable metal, suffice it to say that while the weight of the credible evidence would seem to refute successfully both of these contentions, even so, this fact becomes secondary in the face of our conclusion, already reached, that the Clement patent is invalid because of prior patent disclosure and prior public use.

Thus it becomes unnecessary for us to determine the question of infringement; but it may be said that the construction which we have placed upon the claims in suit and which results in their invalidity because reading upon the prior art, of necessity results in their reading upon defendant's products, and that any other assumed construction, differentiating the claims in suit from the prior art, would likewise of necessity differentiate them from defendant's products, and result in non-infringement on the part of the latter.

In a case of this kind, patentable novelty may reside either in the combination of the elements themselves that constitute the alloy, or in the proportions in which the different elements are combined, or in both. But since metallurgy used in the broader sense as including not only the production of the raw material but its refinement as well is not an exact science, we must recognize that every use of a given patent formula within the specified range of each element will not necessarily produce the alloy contemplated by the patent, which, in legal contemplation, is in fact produced only when one skilled in the art follows the given formula with full regard to the correlation of its several elements in their designated proportions. That is to say, one may pursue the formula, and yet not actually obtain the distinctive product called for by the formula. If so, he does not infringe. By the same reasoning, a given patent formula is not to be treated, in legal contemplation, as having been anticipated if the alleged anticipating product was not in fact recognized by those producing it, as embracing the distinctive characteristics called for by the formula, but was produced accidentally without value or profit to the art. Plaintiffs assert that in the present case, with respect to the question of infringement, they have sustained the burden, which is theirs, of proving that what the defendant has in fact produced is the Clement alloy. For the reasons given, we believe that in this plaintiffs are correct; but as just explained, it becomes unnecessary to enter into a detailed explanation of the similarity of the plaintiffs' and the defendant's products, because we find that defendant has sustained the burden, which rests upon it, of proving that the Clement patent has been anticipated by the Haynes patent, and by prior public use, and is therefore invalid.

Plaintiffs urge the decisions in General Electric Co. v. Hoskins Mfg. Co. (C. C. A.) 224 F. 464, and Pittsburgh Iron & Steel Foundries Co. v. Seaman-Sleeth Co. (C. C. A.) 248 F. 705, in support of their contention that the Clement patent is not anticipated. But both of these cases are clearly distinguishable on their facts. In the first a patent to A. L. Marsh (No. 811,859) for an electric resistance element was upheld, because the court found (page 471 of 224 F.) that Marsh "first disclosed the properties and great advantages of the chromium-nickel alloy as a resistance element. So far as the record shows these had never been suggested in any manner calculated to lead to their application to the uses set out in the patent. Marsh substituted the alloy for the ineffective and practically discarded resistance elements of the prior art. The result was such a remarkable advance upon that prior art as to turn failure into unquestioned success." Similarly, in the other decision, commonly known as the "Adamite" Case, the court found that a patent (to James Ramsey Speer and William L. Forster, No. 1,071,364) for an alloy of iron named "Adamite," which differed from, but had characteristics of, both cast iron and steel, was not anticipated, saying (page 709 of 248 F.): "If any one of the alleged anticipating alloys was Adamite that fact, so far as the record shows, was not known to those who produced it or used it, and not being recognized as a new product with its distinctive characteristics, its production was purely an accident without profit to the art and without value as an anticipation."

We turn now to the second patent in suit, the Hamilton and Evans patent, which relates, not to a product, but to a process; *752 that is, to the production of chromium steel. The single claim (No. 1) in suit reads as follows: "The process of making rust resisting metal which consists in charging a furnace with iron melting the same forming a slag thereon and adding to said slag, ore containing a chromium oxid in the presence of a thermo-reducing agent, whereby the ore is fused, the oxid is reduced and the resulting chromium descends into the molten metal beneath the slag." The process, as more fully disclosed by the specifications, may be described as follows: The chromium oxid ore, together with aluminum or some other thermo-reducing agent, such as silicon or magnesium, is introduced into a suitable slag on top of molten iron in the furnace, usually an electrical one, whereupon the entire body of the ore with its fusable components are fused, the chromium oxid and iron oxide are reduced by the exo-thermic action of the reducing agent, the chromium and iron thus formed descend into the molten metal, while other components of the ore are retained in the slag which not only acts as a flux, promoting the fusion of the ore, but protects the molten metal beneath it from contamination by the carbon electrodes. In other words, Hamilton and Evans claim to have discovered that the proper way to get the chromium into the molten iron is not by putting the chromium ore down into the molten metal, but by putting it into the slag on top of the metal, and having the reaction take place there. That is to say, they found that the best way was not to inject the chromium oxide, that is, the chromium ore, and the silicon directly into the bath of molten metal, because by doing so, the molten iron, for which the silicon has a greater affinity than for chrome oxide, would take up the silicon instead of leaving it free to remove the oxygen from the chrome oxide, but to place the chromium oxide and the silicon, or other thermo-reducing agent, in the slag on top of the metal and cause the reaction to take place there, the oxygen being taken from the chrome oxide unites with the silicon, forms part of the slag and leaves the resultant chrome ore free to flow down into the molten iron below.

We will first consider defendant's claim of invalidity of the Hamilton and Evans patent. Defendant urges that the patent is invalid on the six following grounds: (1) By virtue of section 4887 of the Revised Statutes (35 USCA § 32) because first patented in England on an application filed more than twelve months prior to the United States application; (2) because of prior use by others in the United States; (3) because of prior patents; (4) because of prior publication; (5) because if the claim in suit is construed to cover defendant's operation, it no longer defines the alleged joint invention of Hamilton and Evans; and (6) the entire patent is void for failure to disclaim invalid claims.

Taking up the first ground which defendant alleges for invalidity, namely, that the patent is void under section 4887 of the Revised Statutes (35 USCA § 32), because first patented in England on an application filed more than twelve months prior to the United States application, this section is as follows:

"No person otherwise entitled thereto shall be debarred from receiving a patent for his invention or discovery, nor shall any patent be declared invalid by reason of its having been first patented or caused to be patented by the inventor or his legal representatives or assigns in a foreign country, unless the application for said foreign patent was filed more than twelve months, in cases within the provisions of section 31 of this title, and four months in cases of designs, prior to the filing of the application in this country, in which case no patent shall be granted in this country.

"An application for patent for an invention or discovery or for a design filed in this country by any person who has previously regularly filed an application for a patent for the same invention, discovery, or design in a foreign country which, by treaty, convention, or law, affords similar privileges to citizens of the United States shall have the same force and effect as the same application would have if filed in this country on the date on which the application for patent for the same invention, discovery, or design was first filed in such foreign country, provided the application in this country is filed within twelve months in cases within the provisions of secton 31 of this title, and within four months in cases of designs, from the earliest date on which any such foreign application was filed. But no patent shall be granted on an application for patent for an invention or discovery or a design which had been patented or described in a printed publication in this or any foreign country more than two years before the date of the actual filing of the application in this country, or which had been in public use or on sale in this country for more than two years prior to such filing."

The chronological dates of filing the various Hamilton and Evans applications, both *753 in England and the United States, are as follows: December 4, 1920, application was filed in England for British patent No. 174,271, and concurrently, the provisional specification of this patent as required by the British law was also lodged in the British Patent Office. June 10, 1921, the complete specification of British patent No. 174,271 was filed in the British Patent Office. June 22, 1921, application No. 479,862 was filed in the United States Patent Office with specifications corresponding to the provisional specification of the aforementioned British patent, to which the United States application referred. April 13, 1922, the British patent was sealed, that is, issued by the British Patent Office, and pursuant to British statutes given the date of the original or provisional specification, namely, December 4, 1920. May 23, 1922, application No. 563,020 for patent No. 1,432,289, that is, the patent in suit, was filed in the United States Patent Office, reciting the filing in Great Britain of applications of June 10, 1921, and March 28, 1922, the former date being that of filing the complete specification of British patent No. 174,271 in the British Patent Office, and the latter date that of filing there of another provisional specification which, however, was abandoned and which is not relevant to the present proceeding. Lastly, on October 17, 1922, United States Patent Office issued patent No. 1,432,289 in suit.

Defendant contends, and it is conceded, that under the British practice the only application cognizable as such is the application filed December 4, 1920, that is, the application with provisional specification, and that such date is the one under the English law from which the monopoly of the patent begins to run. Therefore, it is contended that since the United States application No. 479,682, filed June 22, 1921, was abandoned, which is admitted, and since application No. 563,020, upon which the patent in suit was issued, was not filed until May 23, 1922, filing of the latter was too late, thereby rendering the patent in suit void. Plaintiffs, however, reply by saying that the invention covered by the provisional specification, and the one covered by the complete specification filed in the British office, are distinct, and that the British application should not, in interpreting our own statute, be treated as having been filed on December 4, 1920, but rather on June 10, 1921, the date of the filing of the complete British specification, because on the former date the invention covered by the patent in suit had not been conceived, and was not conceived until shortly before the latter date. In addition, plaintiffs say that even if the earlier date, that is, December 4, 1920, is controlling, nevertheless the application filed on May 23, 1922, upon which the patent in suit was eventually issued, should be treated as a continuation of the application filed in the United States Patent Office on June 22, 1921. Plaintiffs further claim that this earlier application was not in fact abandoned until after the filing on May 23, 1922, of the second application, because although the first application was rejected on May 22, 1922, it was still in full force and effect on May 23, 1922, since the inventors were not required to reply to the Patent Office's action of rejection for six months thereafter.

We find plaintiffs' main contention to be correct. It is clear that the British provisional specification did not describe the presence of any slag upon the molten steel bath, upon which slag the mixture of ore and reducing agent is charged, or any other means to assure reduction of the chromium into the bath without incorporating excess of the reducing agent in it at the same time, in accordance with the invention of the patent here in suit. It is admitted that when the British provisional application was filed this process was not known. It is further admitted that it was not known until about the time for filing the complete British specification, namely, June 10, 1921. Hamilton himself testified: "When we filed the provisional we did not mention slag at all, but by the time we were ready for our final, we covered the operation in molten slag. I think that the provisional does not mention slag at all if I remember rightly." It therefore necessarily follows that on May 23, 1922, when the application for the patent in suit was filed, no application for such patent had been filed in the British office more than twelve months previously.

We agree with the construction placed upon the British law by the Court of Appeals for the District of Columbia, namely, that under the British law the application dates from the time of its original filing, simultaneously with the filing of the provisional specification. In re Bastain & Saulsbury, 44 App. D. C. 425; In re Swinburne, 19 App. D. C. 565. But we cannot concede that the construction placed by the Court of Appeals of the District of Columbia upon section 4887 of the Revised Statutes is correct, to the extent that the bar of the statute begins to run from the date of the filing of a foreign provisional specification which ripens into a patent disclosing the subject-matter of the claim upon which a domestic patent is sought, if applied to the present facts. The fallacy of defendant's *754 argument in attempting to apply this rule to the present case lies in the fact that we are here dealing with a situation where the invention described in the United States patent had not been conceived when the British provisional specification was filed. If the District of Columbia decisions, which are not binding upon this court, are properly to be construed as applicable to the situation here presented, we are not inclined to follow them.

An English preliminary or provisional specification might well be so vague and incomplete that its filing could in no way amount to what we, under the American law, understand to be an "application." An English final or complete specification might well be so different from the preliminary specification as to be considered under our law a new "application," and still be, under the English procedure, merely a completing of the record. See English Patent Act, § 42. In this latter situation, the process or device may be described for the first time in the final specification with such definiteness as to make it patentable in this country. Then, since filing this specification would be, under the English law, the proper formal step toward securing a patent, it would not be unreasonable to treat such filing as an "application," in the sense that word is used in Rev. St. § 4887. Such is the situation in the present case. The contrary construction, urged by defendant, is, we feel, unreasonably technical and harsh, and not consonant with the underlying spirit of our patent laws. Therefore, the fact that Hamilton and Evans, in applying for similar patents in other foreign countries, filed their applications within a year of December 4, 1920, does not, as defendant contends, indicate fraud on their part with respect to their action before the United States Patent Office, or determine their rights before that office.

Lastly, it is possible to say that the first United States application was not in fact abandoned but continued by the second. But it is not necessary so to decide, we preferring to rest our decision on this point upon the broader ground just stated.

Turning to the second ground urged for the invalidity of the patent, namely, prior use by others in the United States, we find a great mass of testimony introduced by the defendant which may be summarized as being fragmentary and unconvincing. The prior uses primarily relied upon are those by the Cleveland Brass Manufacturing Company, in its manufacture of an alloy known as "Clebrium," and of the Electro-Alloys Company of Elyria, Ohio, in the manufacture of an alloy known as "Therm-alloy." The testimony of these uses is largely by deposition and with respect to operations that occurred some twelve or fourteen years ago, and is uncorroborated by any satisfactory written records. But even if full credence be attached to the testimony, it has no probative value because the operations referred to do not in fact disclose the Hamilton and Evans process as described in the specifications and in claim 1 in suit, because they did not constitute a process for making a rust-resisting metal as therein defined, but rather related to a different process; that is, one for the derivation of chromium from ferro-chromium. That is to say, it appears from this testimony that before the chrome ore was added in both the Cleveland and Elyria operations, the introduction of chromium to make a heat-resisting metal had been substantially finished, for the metal already contained in each case sufficient chromium for that purpose, and therefore the charge of chrome ore and ferro-silicon at the end of the operation, described as the "last shot," may properly be considered as a mere subordinate feature in the operation. Thus, it may be said with respect to this part of the testimony as a whole, as was said by the Supreme Court in Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45, 60, 43 S. Ct. 322, 327, 67 L. Ed. 523: "The temptation to remember in such cases and the ease with which honest witnesses can convince themselves after many years of having had a conception at the basis of a valuable patent, are well known in this branch of law, and have properly led to a rule that evidence to prove prior discovery must be clear and satisfactory. Barbed Wire Patent Case, 143 U.S. 275, 284, 12 S. Ct. 443, 36 L. Ed. 154; Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L. Ed. 1177."

Partially for the same reason, we are disposed to give little weight to the additional prior art evidence introduced by the defendant to the effect that at several steel plants, so called "High-Speed Steel," containing a high percentage of tungsten, was made in full conformity with the Hamilton and Evans process. We concur in plaintiffs' contention that none of these practices is conclusive with respect to the validity of claim 1 of the Hamilton and Evans patent, for the additional and even more important reason that the process employed was either radically different (Kissock method); the chrome or tungsten ore was added after the melt-down (Ludlum Steel Company method); or none of the *755 chromium contained was derived from chrome ore (Firth-Sterling Steel Company and Le Moyne Steel Company methods). Thus the voluminous testimony with respect to what happened at the Firth-Sterling Steel Company, the Le Moyne Steel Company, the Ludlum Steel Company, or under the Kissock process, does not require detailed analysis.

We next turn to a consideration of prior patents. Here defendant is upon firmer ground than where it stood with respect to prior use. Six prior patents are relied upon by defendant, one of which, that to Price, No. 865,609, application filed November 14, 1905, issued September 10, 1907, we believe covers a process anticipatory in all material features of the Hamilton and Evans process and thereby renders the latter invalid.

In the Price patent, No. 865,609, the process is stated as being "designed for the production of low-carbon ferrochromium, ferromanganese, ferrotitanium ferrovanadium and similar alloys, and contemplates the use of ferrosilicon as a reducing agent. It is possible to electrically produce this silicid with a silicon content of fifty percent and upwards and very low in carbon." It contains seventeen claims, nine of which relate to the specific process of producing ferro-chromium and the remaining eight relate to the process of producing low-carbon ferro-alloys. As substantially typical of these latter claims, claims 3 and 8 may be quoted. They are as follows:

"3. The process of producing low-carbon ferro-alloys, which consists in smelting a charge containing ferrosilicon and a compound of a metal reducible by silicon and alloyable with iron, by passing an electric current through the charge, acting as a resistance-conductor, and thereby heating the charge to the temperature requisite for reduction, as set forth."

"8. The process of producing low-carbon ferro-alloys, which consists in smelting a charge containing ferrosilicon, an oxidized compound of a metal reducible by silicon and alloyable with iron, and a basic flux, by passing an electric current through the charge, acting as a resistance conductor, thereby heating the charge to the temperature requisite for reduction, withdrawing the slag and ferro-alloy from the furnace at different levels, and supplying the charge-mixture as required, as set forth."

It must be conceded that by reason of this Price patent, at the time Hamilton and Evans obtained their patent it was old in the art of smelting refractory ores and producing low-carbon ferro-alloys, to have a molten bath containing iron and chromium and on top of that a slag, and on top of the slag, ferrosilicon chromite ore, which were reduced by an electric are so that as the chromium and iron were reduced out of the ore, they trickled through the slag into the bath beneath. This patent to Price was not cited by the examiner during the prosecution of the Hamilton and Evans application. Plaintiffs seek to differentiate Hamilton and Evans from the Price patent on the ground that although Hamilton and Evans specifically recite (page 2, line 6) that their method is broad enough to cover the production of ferro-chromium as well as numerous other alloys, some referred to, others not, the manufacture of ferro-chromium is not in any sense the manufacture of rust resisting metal, which alone is the subject-matter of claim 1 of the Hamilton and Evans patent, which alone is here in suit, and which defines a method for the manufacture of steel. In other words, plaintiffs stress the fact that ferro-alloys are merely the first products of smelting raw materials, whereas alloying steels are the products of melting down various metals and carrying out careful refining operations. They point out that in the manufacture of ferro-chromium, a very high temperature is employed, considerably higher than that employed in the manufacture of steel; also, that cold ferro-chromium is very coarsely grained. However, it is also true, as substantiated by Dr. Waterhouse, the plaintiffs' principal witness, that ferro-chromium is a rust resisting metallic alloy. And while it is true each claim of a patent is to be treated as an independent invention, and one claim may be infringed, others not (Leeds & Catlin v. Victor Talking Machine Co., 213 U.S. 301, 29 S. Ct. 495, 53 L. Ed. 805), where a prior process patent groups a number of alloys as coming within the contemplation of the specific process covered by the patent, such prior patent must be treated as anticipating the same process in a subsequent patent in so far as the latter relates to the production of any such alloys. Plaintiffs contend that it is of no consequence whatever to the issue now under review, namely, the validity of claim 1 of Hamilton and Evans patent, that some of the other claims of this patent are broad enough to cover the doing of what these patentees have admitted may be done and which would fall within the broad classification of the Price patent. However, this argument is not convincing. See Kissock *756 v. Duquesne Steel Foundry Co. (C. C. A.) 37 F.(2d) 249.

In view of the aforegoing, it becomes unnecessary to dwell upon the remaining five patents relied upon by the defendant. Suffice it to remark that the second Price patent, No. 995,481, application filed June 26, 1907, issued June 20, 1911, calls for the use of carbon as the reducing agent, and for this reason, if not on other grounds also, the patent is distinguishable from Hamilton and Evans.

Next, with respect to the patent to Baraduc-Muller, patent No. 933,357, application filed June 29, 1908, issued September 7, 1909. This invention is defined as relating "to a process of purefying metals and imparting thereto special properties, and more particularly to the production of steels of all qualities" obviously a most ambitious and comprehensive assertion. Since this patent does not describe the manufacture of any type of rust-resisting stainless iron or stainless steel, and, furthermore, since the process specified is quite different from that of Hamilton and Evans, nothing more need be said to show its dissimilarity.

Turning next to patent to Saklatwalla, No. 1,119,643, application filed April 30, 1913, patented December 1, 1914, suffice it to note that the reaction under the process there defined takes place at the surface of the metal bath underneath the slag, and is therefore exactly the reverse of the Hamilton and Evans process.

Turning to the patent to Beneker, No. 1,431,621, application filed March 1, 1921, issued October 10, 1922, we find that patent is for a method of manufacturing steel and more particularly manganese steel. Suffice it to say that the operation appears to be entirely unadapted to the manufacture of stainless iron or steel even though chromium may be used as the alloying element, and, indeed, no claim is made that it is so adapted.

Lastly, coming to the patent to Goldschmidt and Weil, No. 1,235,969, application filed January 20, 1912, issued August 7, 1917, it is sufficient to point out that the process there involved makes no provision for the slag taking part in the reaction which is a very essential part, as we have seen, of the Hamilton and Evans process. For this reason, if not for other reasons, the two processes are distinguishable.

In view of our conclusion that the Hamilton and Evans patent is invalid because of the Price patent, No. 865,609, it becomes unnecessary to dwell upon the three additional grounds for invalidity of the Hamilton and Evans patent urged by defendant which we have not yet considered, namely: (a) Anticipation by prior publication of Bleecker and Morrison; (b) absence of joint invention, namely, invalidity because of use by Hamilton alone before meeting Evans; and (c) failure to disclaim invalid claims. We wish to make it clear, however, that we do not feel that the evidence produced supports the defendant in any one of these contentions.

Having found the Hamilton and Evans patent invalid, it becomes unnecessary to discuss the question of infringement, although we are satisfied from the evidence that the plaintiffs have sustained the burden of proving that the process of manufacture followed by the defendant under the Wild patent, No. 1,586,591, is in all material respects a duplication of the Hamilton and Evans process.

As was said fifteen years ago in the Adamite Case, supra (page 706 of 248 F.), thousands of tests and experiments, and manifold achievements, had already made "the metallurgy of iron and steel a broad and highly developed field. A claim to monopoly of any part of that field by one entering it at this late day can be sustained only by clear proof of discovery of something there not before found, of an invention of something not before there." (Italics inserted.)

For the reasons given, we find that as to all of the claims here in suit, namely, claims 10 and 11 of the Clement, and claim 1 of the Hamilton and Evans patent, defendant has sustained the burden of proving that they are invalid, and that therefore the plaintiffs are not entitled to any part of the relief sought by their bill of complaint, but that, on the contrary, the bill must be dismissed. An order will be signed in accordance with this opinion.