Fried. Krupp Aktien-Gesellschaft v. Midvale Steel Co.

BUFFINGTON, Circuit Judge.

These are four bills in equity brought in’the Circuit Court of the United States for the Eastern District of Pennsylvania by Fried. Krupp Aktien-Gesellschaft, hereinafter called Krupp Company, a corporation of the empire of Germany, against the Midvale Steel Company, hereinafter called Midvale Company, a corporation of the commonwealth of Pennsylvania.

The Krupp Company is the owner of American patent No. 534,178, granted February 12, 1895, for a process of manufacturing armor-plates, on application filed January 24, 1893, to Albert Schmitz and Emil Ehrensberger, residents of the German Empire. This American patent, which is referred to below as the grain-refining patent, is one of a.series of patents for the same alleged invention granted abroad as follows: In Germany, patents Nos. 72,547 and 74,242, dated November 16, 1892; in France, No." 226,286, dated December 9, 1892; in Belgium, No. 102,523, dated December 10, 1892; in Great Britain, No. 23,228, dated December 16, 1892; in Spain, No. 14,054, dated February 9, 1893; and in Italy, No. 33,126, dated December 31, 1892. By one of its bills the Krupp Company charges the Midvale Company with infringing the fifth claim thereof. The Krupp Company is also the owner of American patent No. 642,926, granted February 6, 1900, to said Albert Schmitz for a process of improving the quality of steel armor-plates. It is referred to below as the “fibering” patent. In a second bill the Krupp Company charges the Midvale Company with ■ infringing the first claim thereof. The Krupp Company also owns American patent No. 651,965, granted June 19, 1900, to said Emil Ehrensberger for a process of hardening and toughening armor-plate, and in a third bill it charges the Midvale Company with infringement of the first and third claims thereof. Lastly, the Krupp Company owns American patent No. 653,413, granted July 10, 1900, to said Emil Ehrensberger for a process of improving armor-plate. In a fourth bill it charges the Midvale Company with infringing its first and- third claims-. These last two patents are referred to below as the differential heating patents.

The four cases were heard as one by the court below. There was a large amount of expert testimony, but that court, without entering into a discussion of that portion' of the testimony, confined itself to a consideration of the two fact witnesses who testified to what the Mid-vale Company actually did, as seen by them, in making armor-plate, and reached the conclusion that infringement by the Midvale Company was not proven. In that regard its opinion held:

“Undoubtedly the plaintiff was obliged to establish infringement by the weight of the evidence; but upon this vital point the record contains comparatively little, and in my opinion such evidence as was presented is not sufficiently satisfactory either in quality or amount. For this reason, without touching the other questions involved, the plaintiff must fail.”

In accordance therewith that court entered decrees dismissing said bills for noninfringement. Thereupon the Krupp Company took the *591present four appeals from such decrees. These appeals were heard as one by this court.

The question of infringement being one of fact, we might be justified in following, in the absence of manifest error, the finding of the court below, for, as recently said by the Supreme Court of the United States in Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U. S. 405, 28 Sup. Ct. 748, 52 L. Ed. 1122:

“Tile lower courts, therefore, round that the invention was a broad one, and that the machine used by the Continental Company was an infringement. And these were questions of fact upon which, both of the courts concurring, their findings will not be disturbed, unless clearly wrong.”

But while we might adopt that course here, and while, after full consideration, we are satisfied the case could rest on the basic question of noninfringement which the learned judge below so tersely and pointedly set forth, yet, in view of the importance of armor-plate to several governments, of the fact that numerous companies both in this country and abroad who are not parties to this bill are interested in the subject-matter of this case, and hearing in mind also the industry, research, and zeal of counsel displayed in taking proofs and the argument of these cases, we have felt impelled to record our views more at length than we would otherwise' do and respond to that reasonable -expectation on counsel's part of a full discussion by this court of a -case wherein, to use their own words, “the questions involved are of great importance, not alone to the art, but to the armament of the nations, and, in a broad sense, to the times in which we are living.’’

[ 1 ] We deem it proper, however, to say for the guidance of patent practitioners in this circuit that it should be borne in mind that infringement is not only a question of fact, but is a tort or wrong, the burden of establishing which, as in all torts, clearly rests on those who charge such wrong. The absence of actual fact proof is not met by the presence of expert speculations no matter how voluminous. In this particular case the whole superstructure of the vast mass of .expert testimony, in the last analysis, depends on what the Midvale Company did when making armor-plate as testified to by Leonard and Ross. They were the only witnesses who saw and testified thereto, and, when the judge below became convinced that these two witnesses did not. prove facts which showed infringement, he rested, and could rightfully rest, his decision on that ground. That such a course has commended itself to the master minds of the law is illustrated by what was said of the late Lord Chancellor Halsbury, that “he had exercised a remarkable influence over the judicial temper of the House of Lords by always acting on the maxim that you must first settle what are the real facts of the case before you begin to apply the law.”

The genius of John Ericsson, a native of Sweden, embodied in the original iron-clad Monitor of his adopted country, revolutionized the naval equipment of the world. The Monitor’s battle-tried plates taught the shipbuilder the possibility of a vessel resisting and countervailing the impact of a projectile. From that time the contest in naval construction and equipment became one between the makers of projectiles and the makers of armor-plate, for, as the projectile gained in *592penetrating power, — armor-plate had to increase in resistance efficiency. Thus the two arts supplemented and instructed each.other. At first the armor maker followed the crude pathway of mere bulk, and thickened his iron plates. To such an extreme did this go that in 1871 the British armor-clad Inflexible carried a 24-inch iron armor belt backed'by 11 inches of wood. This armor weight made vessels practically unmanageable; and it was soon found that even 10 and 12 inch iron-clads were so restricted in fuel capacit3r, to say nothing of quarters and supplies for crews, that their steaming radius was shortened, and their protected area restricted to narrow' limits. Moreover, the increased penetrating power of projectiles, due to hardened steel and more powerful explosives, decreased the relative defensive capacity of iron plates. It soon, therefore, became apparent that the vital qualities in a plate were differential strata, viz., a correspondingly hardened face to counteract the hardened projectile and a toughened back wdiich would not only afford cohesion and support to the hardened front of the plate, but would serve to cushion or distribute the ballistic force of the projectile. These facts are summarized by one of complainant’s witnesses, Prof. Howe, who says:

“In order that the plate shall remain intact, its rear must be ductile; that is to say, capable of undergoing a gieat deal of deformation without cracking. The force of the blow given by the projectile is so enormous that, unless the plate is made extremely thick, it must either bulge or break. It must be made no thicker than is absolutely necessary, so that it may weigh down the vessel as little as possible, and thus may give the vessel the-maximum possible coal-carrying capacity, and the maximum possible steaming radius. To break would be fatal, but to bulge under the blow of the projectile does no harm. Therefore the back of the plate should be of material which will neither crack nor break under the impact of the projectile, and therefore must be of ductile material which is capable of bending without breaking or cracking. In steel, in general, hardness and ductility exclude each other. Those steels which are ductile are not hard; and those steels -which are hard are not ductile. Hence conies the necessity of having a radically different set of properties in the face of the plate from those in the back of the plate. The face must be intensely hard in order to arrest and shatter the projectile, and this hardness almost necessarily implies that it shall not be ductile or capable of bending, and, indeed, the lack of ductility has. the great advantage of causing flaking rather than puncture. The back, on the other hand, must be. ductile, lest it crack, and this ductility almost necessarily prevents its being hard. Thus the plate must needs be hard, and hence cannot be ductile in its face, and yet must be ductile, and hence cannot be hard in its back. The successive developments of armor are in the way of perfecting this combination of face hardness with ductility of the back.”-

Solid steel plates of uniform character were also tried, but did not prove satisfactory; for,-if the plates as a whole were made hard, they became nonductile, and, if made ductile, they ceased to be hard. So, while the desiderata of steel plate were recognized by armor makers, no one solved the problem, and the difficulty and' desirability of its solution are evidenced by the expedients used. For example, we find a compound, welded plate, was made, of which that of the Armstrongs of England is a type. In it an impact plate of hardened steel was welded to a tough, wrought-iron back. But, although these were extensively used, they were not satisfactory. The relatively thin steel plate was shattered by impact, the tough back having no cohesive ef*593feet on the hard, but brittle, front to prevent cracking. Moreover, the blow tended to break the nonhomogeneous plate at the weld. These objections were to some extent overcome by Schneider & C.ie at L,e Creusot works in France, where a widely-used, homogenoeus steel plate was made by secret process. So also from these works came the important metallurgical discovery that, by alloying with nickel, the resisting power of a steel plate was not only greatly increased, but that while undergoing sudden cooling, which was a hardening process in manufacturing, the plate was less liable to crack. So important was this discovery considered, as appears by the report of a board of naval officers appointed under the act of Congress of March 3, 1905, and given in evidence by complainant, that the United States government paid a large sum for the right to nickel its armor. It remained, however, for one Harvey, to whom American patent No. 460,262, for decrementally-hardened armor-plate, was granted September 29, 1891, to produce the first satisfactory homogeneous steel plate by a process which the Supreme Court of the United States (United States v. Harvey, 196 U. S. 319, 25 Sup. Ct. 243 [49 L. Ed. 492]) described as one “which has revolutionized the naval armor of the world,” and of which the United States naval board in its report (cited above), after referring to the Ee Creusot invention of nickeling, says:

“The next and even more radical advance in armor-making was the development of the ‘Harvey Process’ of face hardening. This process may fairly be said to have revolutionized the art of armor making.”

Generally speaking, Harvey’s process consisted in supercarburizing .or'“carbon-feeding” a nickel steel plate which was low in carbon by subj ecting its impact face to contact with solid carbon for several days under high heat. The result of such “carbon-feeding” process was that the impact face absorbed enough carbon to make it extremely hard when treated by the hardening process, while the body of the plate gradually diminished in carbon and hardness toward the back. The back, being thus low in carbon, retained its initial soft condition. The result was, as in effect conceded by Prof. Howe, the solution of the problem of producing a homogeneous steel armor plate. His words are:

“The Harvey plate was a great improvement. Harvey enriched the face of the plate in carbon by heating it for a long time in contact with charcoal. He then hardened the face which had finis "been enriched in carbon. He then had a high carbon hardened face which was glass-hard because of its high carbon integrally united to a low carbon back, which was ductile because of its low carbon, and. moreover, lie arranged matters so that the carbon shaded off from high to low gradually. The result of this integral union and of the gradual shading off of the carbon and with it the hardness from the face backwards was that the tendency of the hard and soft parts to separate or flake apart was practically overcome.”

Without committing ourselves on the question, which is not here involved, of the validity of Harvey’s patent, the fact is certain that, in view of his process, nothing that has been done since in the art can be regarded as of pioneer character. The most that can be said is that the subsequent development has been in the line of mere improvement in treatment. The basic feature of Harvey's process was in the mak*594ing of a homogeneous steel plate with the armor desiderata of a hard impact face and a ductile, sustaining back.

[2] In this highly effective state of the art the four patents, which we will consider seriatim, were applied for. These cases being of deep concern, as we have seen, to foreign licensees and governments, we take occasion to briefly refer to the nature of an American patent. Tersely stated, an American patent is a written contract between an inventor and the government. This contract consists of mutual, interrelated considerations moving from each party to the other for such contract. The consideration given on the part of the inventor to the government is the disclosure of his invention in such plain and full terms that any one skilled in the art to which it appertains may practice it.' The consideration on the part of the government given to the patentee for such disclosure is a monopoly for 17 years of the invention disclosed to the extent of the claims allowed in the patent. The right to a patent is statutory, and is provided for by Rev. St. 4888 (U. S. Comp. St. 1901, p. 3383), which is:

“Before any inventor or discoverer shall receive a patent for his invention or discovery, he shall make application therefor, in writing, to the Commissioner of Patents, and shall file in the Patent Office a written description of the same, and of the manner and process of making, constructing, compounding, and using it, in such full, clear, concise and exact terms as to enable any person skilled in the art or science to which it appertains, or with which it is most nearly connected, to make, construct, compound, and use the same; and in case of a machine, lie shall explain the principle thereof, and the best mode in which he has contemplated applying that principle, so as to distinguish it from other inventions; and he shall particularly point out and distinctly claim the part, improvement, or combination which he claims as his invention or discovery.”

[3] Taking up, then, the patent to Schmitz and Ehrensberger, No. 534,178, issued February 12, 1895, and which forms the subject-matter of the first bill, we inquire what was the invention therein disclosed. This patent is referred to in the record as the “grain refining patent.” But, before inquiring what the patent is for, it is well to understand clearly for what it is not, namely, that it is not and does not purport to be for a product. In other words, it is a process, and not a product, patent. It is, as the patent states, for a “process of manufacturing armor plates.” This distinction between process and product patents must be kept in view in considering patents, such as are here involved, otherwise we are apt to conclude from the mere fact that similar products are made by two different persons that one is infringing the other’s rights. On the contrary, in such cases, the real test of infringement is not identity of a product which is not patented, but identity of patented process in producing an unpatented product. At this point it will be noted that both the practice of carbon-feeding steel and the hardening and grain refining of such carbon-fed steel were well known prior to this patent. As to carbon-feeding armor-plate, the patent to Harvey, No. 460,262, of September 29, 1891, states:

“This invention embraces a method of facilitating the transformation of homogeneous low steel armor-plates into plates which present upon the side intended to receive the impact of projectiles a stratum of tenacious steel of *595heterogeneous crystalline structure highly and uniformly carburized and excessively hard upon its exposed surface, and less and less carburized .and gradually diminishing in hardness as the depth from said surface increases. The required transformation is effected by inclosing the low steel plate between a mass of noncarbonaeeous granular material on one side and a mass of granular carbonaceous material packed firmly against, the other side in a compartment erected within the heating chamber of a suitable furnace and in then raising the heat of said heating chamber and maintaining it at a temperature above the melting point of cast iron for a period of time sufficient to effect the desired increase in the tenacity of the steel and the supercarburization to the desired extent and depth of the side of the plate against which the granular carbonaceous material is being constantly pressed. The plate is subsequently removed from the furnace and chilled h.v immersion in a cold bath or otherwise, as hereinafter set forth, whereby its superenrburized side is hardened. The heat to which the plate is subjected during the described treatment is so intense that the plate would be melted but for the granular materials surrounding it, h.v which all parts of it are protected from the air. The continuous firm compression of the carbonaceous material against the plate during the entire treatment secures the perfect contact, of the carbonaceous material with all portions of the adjacent side of the plate and promotes the rapid and uniform supercarbnrization thereof. * * * The armor-plate having been formed of the desired size and shape from a comparatively low steel, such as Bessemer steel or open-hearth steel, containing, say .10 to .35 per cent, of carbon, is laid, preferably, flatwise upon a bed of finely-powdered dry clay or sand deposited upon the bottom of a fire-brick cell or compartment erected within the heating chamber of a suitable furnace. The. plate may be so embedded that its upper surface is in the same plane with the upper surface of those portions of the bed of clay or sand which adjoin the sides and ends of the plate, or the plate may, if (iesired, be allowed to project to a greater or less distance ahoye the surface of the clay or sand. In either case the treating-compartment is then partially filled up with granular carbonaceous material, which, having been rammed down upon the plate, is covered with a stratum of sand, upon which there is laid a covering of heavy fire bricks. The furnace is then raised to an intense heat, which is kept up for such a period of time as may be required for the absorption by the metal adjoining the upper surface of the plate of, say, an additional one per cent, (more or less) of carbon, or, in other words, the quantity of carbon, in addition to that originally present, which may he necessary to enable the said metal to acquire the capacity of hardening to the desired degree.”

But not only was the general principle of carbon-feeding armor-plates here shown old, but so also was treatment of them by applying solid carbon in a “compartment erected within the heating chamber of a suitable furnace,” as the above extract shows. . Moreover, Harvey showed such compartment was air-tight. Thus he says:

“The heat to which the plate is subjected during the described treatment is so intense that the plate would be melted but for the granular materials surrounding it, by which all parts of it are protected from the air.”

So, also, it will he noted that the patentees themselves, as quoted below, concede that the use of hydro-carbon gases for carbon-feeding iron was known.

It will also be seen that grain refining was a well-known prior practice in the steel art. Complainant’s witness Langley, in summarizing the general prior practice of steel-making, says:

“High steel [that is, steel containing more than .5 per cent, carbon] possesses a very remarkable and substantially unique property of acquiring an Intense degree of hardness, so that it cannot be scratched with a file or *596knife, this condition being given to it by first heating it to a temperature above a dull red heat, and then cooling it suddenly by bringing it in contact with cold water or oil, and the intensity of this hardness is increased appreciably in proportion to the suddenness of the cooling. *' * * If steel is hardened at only a few degrees above this critical point, it requires an exceedingly fine crystalline and velvety grain. It is now intensely hard and possesses its maximum of strength, although it is still very brittle as compared to its condition before hardening.”

[4] In this state of the art the patentees applied for their patent, and to it we turn to find what disclosure they made of a novel and inventive character in this already well-developed art. The nature of their disclosure is all set forth in plain terms in their specification, and we may here remark that it is only where ' a court. requires the explanation of technical terms or the language of any particular art to enable it to understand a specification that the evidence of those expert in the art is pertinent. The province of such expert testimony is to remove uncertainty where the terms used are obscure, and is not to create uncertainty where the language is plain.. In this specification the words and terms are intelligible, and, viewed from the standpoint of their common, ordinary meaning, .a study of this patent satisfies us that all that was disclosed by these patentees were processes for treating armor plates by carbon feeding and grain refining by heating, by the use of particular appliances, and suddenly cooling.

[5] These processes were, first, by the use of hydro-carbon gas; second, by solid carbon; and, third, means by which these processes were used, to wit, a particular form of an air-tight, intrafurnace compartment, placed within an ordinary heating furnace. It will, of course, be noted that the case before us does not involve the first or hydro-carbon gas process, and we have therefore no present concern with, and, of course, express no opinion upon, the claims granted thereon. They are only here referred to because a due consideration of that feature is, in our view, essential to an understanding of the patent as a whole. Turning our attention then to the process for utilizing hydro-carbon gases in the intra furnace chamber shown, we find the patentees say:

“It is known that iron deficient in carbon can be quickly carbureted by passing hydro-carbon gases over the same, and, according to our invention, we utilize this knowledge as hereinafter described for the purpose of giving a hard surface to one face of otherwise soft armor plates of all kinds. Our invention will be readily understood from the following description having reference to the accompanying drawings: Figure 1 (shown below) is a vertical, longitudinal section of a furnace adapted for carrying out our process. Fig. 2 (shown below) is a transverse section, and Fig. 3 is a horizontal section of the same furnace. * * * The furnace A is heated by gas which enters through the passage a, while the air is supplied through the passage 6. The products of combustion escape through two passages e that are situated on either side of the door d. The hearth B of the furnace is capable of being drawn out from the latter, say, on rails as shown. The first plate 0 to be carbureted uppermost, upon supports or pillars e on the hearth of the furnace, the said hearth having been drawn out of the said furnace, while a second plate D is laid with the face to be carbonized toward the first plate and is held at a small distance from the latter by means of intermediate pieces of iron f, stone or other suitable materials. When so placed both ifiates have walls g g h i built round them, so that the space between the two armor-plates is completely closed with the exception of two holes K in the *597wall h and two largor holes l in the wall i, while the furnace gases can pass freely above and below ¡lie exterior surfaces of the respective: plates, the gases passing under the plaie C through openings m, formed in the walls h and After the two plates have been built up in this manner on the hearth, the latter is pushed into the furnace and the latter is heated, until the plates are about the melting temperature of copper, whereupon pipes n and o. the positions of which are indicated at Fig. i. are passed through two openings in the furnace door into the holes l in the wall i, and hydrocarbon, illuminating gas, parafme vapor, petroleum gases or the like, are passed through said pipes in between the plates. The surfaces of the plates become carbureted by the action of these gases in a comparatively short time, while the used gases, deprived of a large proportion of their carbon, pass through the openings 7,' into the wail h in the furnace and burn there. As soon as the carburization of the plates has reached the desired degree, the hearth is run out and the plates are hardened by dipping into or by spraying the part to be hardened, with water or oil.”

Upon this process four claims were allowed, and these we quote below to show that the hydro-carbon gas treatment was a most substantial part of the disclosure, and that both specification and claims showed a process which made use of an intrafurnace structure. Upon this disclosure four claims, not here in dispute, as stated before, viz.:

“1. The described process of cementation of plates or articles to be carbonized on one side only, said process consisting in forming in a furnace a gas-tight chamber or inclosure having the surface of the plate to be cemented for a portion of the wall thereof, heating the plate and admitting to .said chamber gaseous cementing agents, substantially as set forth.
“2. The hereindescribed process for the manufacture of armor-plates, etchardeued on one side only, consisting in building the same into a furnace with the faces to be hardened opposite to each other and with an intermediate space, heating said plates, then admitting a gaseous hydro-carbon into the inclosed space and discharging the used gases deprived of a large proportion of their carbon.
“3. The hereindescribed process for the manufacture of armor-plates, etc., hardened on one side only, consisting in building the same into a furnace with the faces to be hardened opposite to each other and with an intermediate space, heating said plates, then passing a gaseous hydro-carbon into the inclosed space, and finally hardening, substantially as described.
“4. The hereindescribed process for the manufacture of armor-plates, etc., hardened on one side only, consisting in building the same into a furnace with the faces to be hardened opposite to sach other and with an inclosed intermediate space: heating said plates, then introducing a gaseous hydrocarbon at one side of the intermediate space and discharging it at the other, and finally hardening the carbureted side, substantially as described”

- — were granted covering the use of hydro-carbon gas. Each of said claims had some element descriptive of the intrafurnace chamber, wherein and whereby the plates were subjected to the disclosed process. Thus the first claim says the process consisted “in forming in a furnace a gas-tight chamber or inclosure having the surface of the plate to be cemented for a portion of the wall thereof.” The second “consisting in building the same [the plates] into a furnace * * * then admitting a gaseous hydro-carbon into the inclosed space and discharging the used gases deprived of a large proportion of their carbon,” The third “consisting in building the same [the plates] into a furnace * * * then passing a gaseous hydrocarbon into the inclosed space.” The fourth “consisting in building *598the same [the plates] into a furnace * * * with an inclosed intermediate space; * * * then introducing a gaseous hydro-carbon at one side the intermediate space and discharging it at the other.” Moreover, such an interior structure is shown in those drawings, which are taken from the application:

Thus Fig. 1 shows the end walls of the structure by the letters h, l, and Fig. 2 the side walls by the letter g; and the specification says:

“When so placed both plates have walls g g % i built round them, so that the space between the two armor plates is completely closed with the exception of two holes. * * * The surfaces of the plates become carbureted by the action of these gases in a comparatively short time, while the used gases deprived of a large proportion of their carbon, pass through the openings 7c into the wall h in the furnace and burn there.”

*599Indeed, unless we adopt the view that the particular form of the intrafurnace structure constitutes the novelty in the process disclosed, the patent would be invalid, for, apart from its specific details of such inner chamber, there is nothing to differentiate it generically from the air-tight, carbon-feeding, intra furnace structure which, as we have seen, Harvey showed in his prior patent. It is evident that the gist ■of Schmitz and Ehrensberger’s disclosure lay in this structure in which hydro-carbon gas was used, and which as claimed by them carbureted the plates “in a comparatively, short time.” And this view is emphasized by the fact that, as shown by complainant’s proof, the process used by the Krupp Company at Essen, and which may therefore be taken as embodying the preferable practice of the disclosure, was of hydro-carbon gas in an intrafurnace structure. Thus Mr. Unger of the Carnegie Steel Company, a licensee under these patents, who visited Essen to familiarize himself with the process, says of the Essen practice:

“The plate was then carburized by means of the carbon deposited from ordinary illuminating gas, which was allowed to flow into a chamber between two plates.”

We are therefore warranted by tlie prominence given in the specification, in tlie drawings, the number of claims addressed to that feature, and by the practice 'of the Krupp Company in concluding, as we do, that the use of hydro-carbon gas in the form of furnace shown constituted the principal element of such patent.1

Ilaving disclosed such alleged invention by the use of hydro-carbon gas in an intrafurnace chamber of specified form, the patentees proceed to state (and this is their sole disclosure as to the solid carbon process) as follows:

“The same arrangement as employed, according to our invention, for carbureting with hydro-carbon, may he also employed for carbureting with solid ea rbon or the like, by merely Ailing with the same tlie space between the two armor plates into which the gas would otherwise enter. The holes fc and ?, and the supports between the plates, are dispensed with in this case. Tills method of hardening two plates together with solid carbon is an important Improvement on the heretofore, usual method of carbureting one plate only at a time.”

Now, it is clear beyond contention, and the extract from the specification just made so states, that the fifth claim was meant to embody the same form of intrafurnace appliance used in the hydro-carbon gas process, for the specification says:

“The same arrangement as employed, according to our invention, for carbureting with hydro-carbon, may also be employed for carbureting with solid carbon.”

Indeed, complainant’s proof shows that, when the Carnegie Steel Company used carbonaceous matter instead of hydro-carbon gases, *600it employed an air-tight, intrafurnace chamber. Thus Mr. Unger of that company says of its practice:

“The spaces which may exist between the edges of the plates and the angles is now luted up tight with a mortar made of water and fire clay, the object of this being to prevent as far as possible the access of air to the bed of carbonaceous material during the heating of the plate in the furnace.”

Now, unless this claim is restricted to the use of a chamber such as disclosed in the specification and shown in the drawings and having end walls and side walls, wherein shall we find patentable novelty in the claim? For, if we eliminate that feature and attempt to support the claim otherwise, on what can it rest except the mere mechanical, manufacturing fact that, instead of carbon feeding one piece of armor plate in an air-tight interior furnace structure as Harvey.had done, Schmitz and Ehrensberger discovered that by stacking one plate on another they could carbon feed two plates at a time? And, after all is said, is not that whai claim 5 consists of? As we have seen, the patentees do not attribute any novel result to this process. They only say:

“Tliis method of hardening two plates together with solid carbon is an important improvement on the heretofore usual method of carbureting one pílate only at a time.”

Wherein does the improvement consist? The patentee discloses none save duplication. And such appears to be the conclusion of Prof. Howe, who when testifying in rebuttal for complainant, and after full discussion of this patent in chief, summed up his estimate of this claim by. saying it provides—

“an efficient means for carburizing the face of the plate so as to carburize one or several pairs of plates simultaneously instead of one plate at a time. This same arrangement had the further advantage of making use of' the plates themselves as a muffle or protecting wall to prevent the carbon used for carburizing the plates from being burnt away by the flame of the furnace, and also of making use of the weight of the plate to press the charcoal firmly against both the upper and lower plate, thereby hastening the carburizing, excluding the air, and so lessening the waste of charcoal, and promoting uniformity of carburizing by means of this pressure. * * * The Schmitz and Ehrensberger patent gives useful and economical administrative methods.”

But mere useful and economical administrative methods, however valuable, while they may and usually are incident to invention, do not themselves constitute invention. If stacking one armor plate on another was an inventive act on the part of these patentees, then the practice of the Midvale Company, where six and eight plates were simultaneously stacked and treated, constituted a different invention and one of even higher order. We cannot accede to such reasoning in the domain of patent law. Indeed, to use a very homely illustration, What was this two-plate stacking but sandwiching two plates, so that two,-instead of one, should be fed by the same carbon? The process arid effect of carbon feeding armor plate as a preliminary .step to hardening being well understood and practiced by Harvey and hardening and grain refining being well known, did it involve patentable novelty to duplicate the practice of carbon feeding one armor *601plate at a time by superimposing another plate? It is assumed by the expert quoted above that the element of pressure was involved, but pressure in carbon feeding was not a practice originated by Schmitz and Ehrensberger. They made no such averment in their specification, and Harvey in the extract quoted refers to his practice of having a “mass of granular material packed firmly against the other side,” and to maintaining conditions for a sufficient time “to effect the desired increase in the tenacity of the steel and the supercarburization to the desired extent and depth of the side of the plate against which the granular carbonaceous material is being constantly pressed. * * * The continuous firm compression of the carbonaceous material against the plate during the entire treatment secures the perfect contact of the carbonaceous material with all portions of the adjacent side of the plate and promotes the rapid and uniform supercarburization thereof.” He further says:

‘■The armor-plate having been formed of the desired size and shape from a comparatively low steel, such as Bessemer steel or open-hearth steel, coats filing, say, . 10 to .85 per cent, of carbon, is laid, preferably, fla twise upon a bed of finely-powdered dry clay or sand deposited upon the bed of a fire-brick coll or compartment erected within the heating chamber of a suitable furnace. The plate may be so embedded that its upper surface is in the same plane with the upper surface of those portions of the bed of clay or sand which adjoins the sides and ends of the plate, or the plate may, if desired, be allowed to project to a greater or less distance above the surface of the clay or sand. In either ease the treatment is then partially filled up with granular carbonaceous material, which, having been rammed down upon the plate, is covered with a stratum'of sand, upon which the,re is laid a cover-mu of heavy fire bricks. * * * The more intense the heat the better, and while it will, of course, be understood that the longer the treatment is continued the greater will be the depth la which the carbon penetrates beneath the surface against which the carbonaceous material is packed, it is also to be remarked that the penetration of the carbon Is-greatly facilitated by the continuous firm compression of the carbonaceous material against the plate. * * * A stratum of granular carbonaceous material E rising to a heighth, of say eight inches above the upper surface of the plate D, is tightly rammed down onto the top of the plate and is surmounted by a stratum of, say, two inches of sand P, covered by a layer of heavy fire brick. The stratum, of savA P and the, layer of fire brick ft not only protect the carbonaceous -material from, the fire, but serve lo weight the carbonaceous material down upon the plate."

It will thus be seen that the element of carbon-feeding pressure on which Prof. Howe lays stress as a distinctive advantage of the patent was practiced by Harvey. Consequently the novelty of the fifth claim is reduced to the substitution of a second plate as a weight in place of fire brick. This, it surely must be conceded, was a mere mechanical, administrative practice, and involved nothing patentable. The lower plate of the patentees’ duplication was in no wise differently carbon-fed than in Harvey’s process, and from the standpoint of intensifying such carbon-feeding by pressure the lower plate was in no wise differently affected than by Harvey’s ramming of the carbon, and using layers of sand and heavy fire brick. Mere simultaneous treatment of two objects by the same process previously applied to one is not invention. It therefore follows that claim 5 of this patent is void for lack of patentable novelty.

*602We express no opinion on the question'whether the fifth claim discloses a workable process, and whether the entire patent is void by reason of the expiration of earlier foreign patents for the same alleged invention.

[6] We turn next to patent No. 642,926, to Albert Schmitz, which forms the subject-matter of the second bill, and the first claim of which is alleged to be infringed. This patent is called by complainant’s counsel “the fibering patent,” and it is contended the process therein disclosed is epoch making, and was the first known process to produce what is termed fibering or giving a distinctive tensile structure to armor-plate. This patent was granted February 6, 1900, and is for “a process of improving quality of armor plate.” Assuming, for present purposes, that there is such a thing as a distinctively fibered steel armor-plate, and that such a product was for the first time produced by this patentee’s process, it is not without significance that, so far as the proofs go, the patentee made no effort to secure patent protection for such a product, and during a long and labored effort to obtain a process patent he made no statement that fibered armor plate as a product was original with him. Furthermore, on study of the proceedings during the application for the patent, it will be seen that the process sought to be patented is not, and does not purport, to make armor-plate in toto, but merely avers that the process is a treatment to further improve already completed armor-plate. This the patentee makes clear, saying:

“The process forming the subject of the‘present application has for its object the improvement of the quality of steel armor-plates by subjecting the same after their production by any one of the known processes of manufacture to a series of additional metallurgical operations.”

[7] Now, it is quite clear that armor-plate produced by “any one of the known processes of manufacture” has a front or impact surface and a back or supporting surface, which result from different thermal and cooling treatments, for hardening the face and toughening of back resulting from the respective heating and cooling processes the respective fronts and the backs of the plates have each undergone. Indeed, one of complainant’s experts saj^s that “a piece of steel indelibly records in its internal structure the heat treatment which it has received.” These different heating and chilling processes being so acute as to leave indelible evidence in the structure of a plate produced by any one of the known processes of making armor plate, and the further fact being that in the higher ranges of the different processes the changes in metallic condition are more rapid and radical, it follows that when further and additional thermal and chilling processes are proposed, and especially when such processes are of the thermal heighth where rapid and radical changes result, the bounds of such further processes and the exact nature of such steps should be disclosed, and such definite and accurate directions given as to enable any one skilled in the art to apply the process to the several kinds of previously made armor-plate to which the patentee claimed his process was applicable. In other words, the teaching of the specification and the disclosure by the patentee must be *603such that, after the patent has expired, a user thereof shall not be left to the blind groping of experimental work, but by the plain teaching of the specification be enabled to use the process with certainty. In view of the statutory requirement of clearness in the disclosure — ■ the clearness of the disclosure equally with the novelty and utility of the invention form the consideration for the contract monopoly — we are confronted at the outstart by the sworn statements of two men, highly versed in the steel art, who were called on behalf of the respondents, who say, in effect, that the directions to work the patent are unintelligible to them. Joseph Hartshorne qualifies as a consulting mechanical and metallurgical engineer. His professional equipment, practical work in the art, and the intelligent, fair-minded general character of his testimony impress the court with the reasonableness of his conclusions. In speaking of this patent he testifies:

“As far as I can understand it, it seems to describe an utterly unworkable process, no matter in what way its various descriptions and limitations are taken.”

The other is Albert Sauveur, Professor of Metallurgy at Harvard University, tie testifies:

“Summing up my understanding of this patent, it describes two treatments, namely, A and B, which were old at the time of the application for the patent [and this we may say is conceded] and had been old for many years prior i hereto, and a third treatment, C, which is evidently an inoperative one, disclosing an unworkable process. * * * This is a metallurgical impossibility, and the patentee gives no information as to the_ way it should be performed, and I certainly cannot conceive of any way of performing it.”

Indeed, the respondent’s witnesses do not stand alone in their criticism of this patent for the able expert witness, for the complainant, who was called to testify in the prima facie case, said:

“Much of the language of this patent is a very good illustration of the vague and often unsupported statements which are tolerated in foreign patents, of which this is essentially one, although it has been issued in this country. For example, the statement occurring on page 2, lines 8 and 9, which follows the assertion of the patentee that his process is applicable to various kinds of plates, reads as follows: ‘And also to compound plates which consist only partially of steel.’ This is ludicrously absurd in its indefiniteness. for what the other portion of the plate which is not steel may be is left wholly to conjecture, and it is self-evident that, if the other part of this particular plate was made of copper, of east iron, or a great many other substances, it is self-evident that the result of heating it and treating it as the patent describes would yield a result not characterized by a ‘highly fibrous molecular grain.’ Another example of the loose language of this patent is found on page 2, lines 3 to 55: ‘This new process is particularly suitable for nickel-steel armor plates of various degrees of hardness.’ Under this liberal permission the plate might be made entirely of very low carbon steel, but both the specification and claim. 1 require that the plate shall be hardened. Now, as very low carbon steel cannot be hardened to more than an infinitessimal degree by sudden cooling, it is apparent that the results called for in claim 1 are physically unattainable on such steel.”

Furthermore, referring to treatment 3 of claim 1 quoted above, in which treatment alone the novelty and infringement of the process are averred to lie, he says:

“The language of claim 1 of said Schmitz patent contains the following expression: ‘Below that permitting retroversion of the granular structure *604and allowing the plate to cool slowly.’ Turning now to the specification to find out what the patentee means by the term granular structure, I do not find that this word is used, or that there is any language which clearly and apart from conjecture denotes what he means by granular structure.”

So, also, it is not without significance that the testimony of Mr. Unger of the Carnegie Steel Company, a licensee under this patent, inferentially shows that, although he and his company had before them the teaching of the patent, it did not begin making armor under it until he had visited the Krupp works, and there observed their methods. Under the evidence before us, there are grave grounds to support the contention that the disclosures of the patent are not in such clear terms as to enable persons skilled in the art to use the same, and we might be warranted in holding this patent void for uncertainty. We prefer, however, to assume for present purposes its validity, and determine the case on broader grounds.

In taking up the question of the respondent’s infringement, we are first led to inquire what is the monopoly to which the complainant is entitled. And in doing so we must note, as we did above but with reference to a different matter, that where, as .here, the process which forms its general subject-matter deals with such tremendous heats as are here employed and the opposite and proportionately radical ranges of cooling, and where the plates are so thick that faces and backs may simultaneously have widely different thermal conditions respectively front and back and where the metallic changes are intensified both in character and rapidity when the thermal condition exceeds certain points, it will be apparent that any substantial departure in process, either in the way of heat ranges, rapidity or slowness of cooling, and subjecting the plate to whole or only partial or stratified heating, would, in the just measurement of patent rights, make such variations, presumably, a different, and therefore for the purposes of infringement, a noninfringing, process. Moreover, it will be here noted, as stated above, that, while this process is one for changing the composition of fiber of the plate, the patent is not, as we noted in considering the former patent, for a product, that is a fibered product, and therefore covering any and all fibered armor-plate, but is for a process for obtaining such a product. It therefore follows that, because the Mid-vale Company or any other armor-plate maker makes fibered armor-plate, that product is not proof of any infringing act; for, to visit such maker with infringement, it must be shown by proof of facts that such fibering is done by methods disclosed in the patent and specified in the first claim. To illustrate, Mr. Unger in speaking of the . Krupp-Essen practice shows that that company initially fibered armor-plate by some undisclosed method, in addition to the subsequent fibering which alone is alleged to come within the scope of the patented process. Thus, in speaking of the Essen practice, he says:

“After tlie plate Rad been rolled, it was subjected to a treatment for the purpose of producing a fibrous condition throughout the plate to prevent it from cracking when the plate was allowed to cool and to put it in a soft condition to- permit any preliminary machining before carburizing.”

*605It will here be observed that by this undisclosed process, which, if done by a third person, would not, so far as the proofs go, have infringed the patent the Krupp Company fibered the plate once; so that the mere fact that the Midvale Company’s plate was fibered does not establish infringement. Indeed, as that witness further shows, the fibering of the patented process is an operation subsequent to carbon-feeding and grain-refining, which operation he thus describes:

“The plate was then reheated to a temperature of about 600° to 650° Cent., and then cooled in water, so as lo ay am put the plate in a fibrous condition.”

It will thus appear by complainant’s own showing there is a fibering process in use in armor-plate practice, and that, too, in the making of armor-plate by the Krupp Company, wThich is not shown to be the alleged fibering process of the patent.

Turning now to the specification of Schmitz’s patent, we find he thus describes the process:

“If a steel armor-plate of any composition, as. for example, nickel steel, compound, or homogeneous plates, as hereinafter stated, be heated to a temperature producing a pasty condition permitting a molecular rearrangement, say, for example. 800° Centigrade, and then hardened in this condition by a complete immersion in oil or water or other liquids, the plate will assume a greater or less degree of hardness, with the formation of an extremely fine-grained structure. The temperature required for this purpose is higher the less the amount of carbon contained in the plate, and depends also on the chemical composition of the plate. Now, if this plate thus hardened be again subjected to a glowing heat, which must, however, not be so high as the previous hardening heat, say 600° Centigrade, and particularly not so high as to allow of the possibility of ready molecular rearrangement taking place, and if the plate as soon as it has readied uniformly and with certainty the temperature corresponding to the said glowing heat or is maintained for some hours at such temperature be allowed to cool slowly, the hardness proper of the plate is lost. The plate retains, however, a considerable greater resisting power against shot than the untreated plate, and acquires above all by the highly fibrous molecular grain thus produced (which has an appearance on fracture very similar to velvet with a silky border) the capacity in a very high degree of offering great resistance to fracture when struck by shot. If, now, the plate thus improved be again heated to the last glowing heat or to a lower temperature, which shall not, however, exclude hardening. and the plate be hardened on one face by copious drenching with water or by immersing one face of the plate in water which may also contain any known hardening agents, while the other face of the plate remains out of contact with the water, then an armor-plate is produced having a very hard face for exposure to projectiles, while the mass or main body of the plate has been converted by the preceding operations into a condition that oilers the greatest possible security against fracture combined with increased resistance.”

Upon this disclosure was granted the claim here involved, viz.:

“1. The herein described method for the improvement of armor-plates, consisting in first heating the plate to a temperature permitting molecular rearrangement and consequent conversion upon hardening to a fine, homogeneous grain, and then hardening the entire plate by suddenly cooling, then reheating to a maximum temperature below that permitting retroversion of the granular structure and allowing the plate to cool slowly, then reheating the same to a temperature equal to, or somewhat lower than, the temperature of the second reheating, but not so low as to exclude hardening and then hardening (by suddenly cooling) one surface of the plate only.”

*606The carbon-feeding of an armor-plate leaves the steel with a comparatively large grain. This granular largeness is a source of weakness, for, when struck, the plate lacks the cohesion necessary to prevent cracking. This is clearly stated by Prof. Langley, one of complainant’s experts, who says:

“The effect of long continued high heat on solid steel is invariably detrimental, tending to throw its structure into a coarsely crystalline condition in which its fracture resembles that of cast iron, and is accompanied by great loss of strength and an almost total abolition of the property of toughness. The higher the percentage of carbon, the more detrimental is this change. Now, the necessarily long-continued high heat to insure carburization has unavoidably produced this detrimental change in the plate, which, if it could not be remedied, would render it nearly worthless; for, although the withdrawal of the carburized plate from the furnace and then rapidly chilling It will insure a high degree of hardness to the carburized face, it will not remedy the coarsely crystalline and weak condition brought about by the long-continued high temperature. Subsequent remedial measures must therefore be taken.”

This coarseness of grain the patentee changes by what was old in the steel art, and is known as “grain-refining.” This grain-réfining forms the first step of the claim, viz.:

“Heating the plate to a temperature permitting molecular rearrangement and consequent conversion, upon hardening, to a fine, homogeneous grain.”

It is described in the specification:

“If a steel armor-plate of any composition * * * be heated to a temperature producing a pasty condition permitting a molecular rearrangement, say, for example, 800a Centigrade, and then hardened in this condition by complete immersion in oil, or water or other liquid, the plate will assume a greater or less degree of hardness with the formation of an extremely fine grained structure.”

This “pasty” condition, and the transition -from coarse grain to fine, is, according to Prof. Langley, complainant’s expert, reached at what is known as the “critical point,” which condition and traxrsition he thus describes:

“The obvious meaning of the terxn ‘pasty’ condition permitting a molecular rearrangement clearly refers to the transition in the internal structure of a piece of high carbon steel, which takes place when the texxiperature crosses the ‘critical point’ or ‘recalescent’ point’ as I have explained in my answer to your questioxx 5, for it is well known by recent inxcx’oseopieal researches that the internal structural of such steel undergoes a profound change at the critical point.”

The process of grain refining was well known in the pidor art and one which the respondent was free to use; for, without citing at length the numerous citations from which he draws his opinion, we may say that an examination of them satisfies us that the witness Hartshorne was justified in summarizing the state of the ail by saying:

“It is apparent from the above citation and in my own knowledge that the method of gi’ain-refining by heating above the critical point and then suddenly cooling the steel has been known since 1S7S and xxrior thereto, and that the purpose of this treatment is to make the grain as-fine as possible in order to impart the greatest strength to the steel, and in order to allow the greatest toughness to be imparted to it by a subsequent annealing.”

*607After this is completed, the plate is then hardened, or, as stated in the claim, “then hardening the plate by suddenly cooling.” The second step is described by complainant as fibering the plate, and their contention is that this is a novel, and, as their expert in rebuttal says, an epoch-making, invention. The word “fiber” is not new as applied to metals.2 3

With the consideration due to the conceded ability of complainant's experts and with high regard for the ability of counsel, we have not from a study of the proofs been convinced either that there is such a distinctive structured article as fibered armor-plate, or, if there is, that Schmitz’s process was the first to create fiber therein, or, indeed, that the second claim is anything more than annealing. As we have noted before, the patentee made no averment that he was originating any new metallic product. On the contrary, all he says is that the second step of his process produces a highly fibrous molecular grain “which has an appearance on fracture very similar to velvet with a silky border”; in other words, a grain of a highly, as contrasted with a less highly, developed fibrous character. In fact, his second step is but a species of annealing, as he himself says :

“As will be seen from the preceding, the present process consists essentially in the following three operations: * * * Second, annealing the plates to a temperature which is as high as possible without, however, reaching the temperature at which said capability of ready rearrangement begins.”

[10, 11] It need not be said that the principles and general processes of annealing were well known. Annealing is done in several arts, and is defined as “the process used to render glass, iron, etc., less brittle by allowing them to cool very gradually from a high heat,” or, as stated by Prof. Langley, “the effect of the annealing treatment is to replace the iron-like grains caused by the carbonizing stage by a different grain which is called fibrous”; and “the term ‘fibrous’ is a generic one used in the steel-making art to denote a condition in contrast to the crystalline state.” In the statement of the general properties of steel and the practice of steel-making which Prof. Langley made at the outset of his testimony, he, in effect, describes the method of annealing in order to grain refine and produce a fibrous toughness. After stating the effects of carburization, as quoted in a foregoing extract, he says: *608gree its coarse cast-iron grains changed into a tough fibrous state by a long continued application'of a heat below the hardening point; that is, by a heat ranging between 700° C. as .an upper limit down to about 600° C. as a lower limit. This change to a' fibrous condition applies particularly to low carbon steel, while its effect on high carbon steel, which has been hardened by chilling, is to soften it1 and give to it a fine semicrystalline grain possessing some toughness, but not to the same degree as that of the fibrous condition produced in low carbon steel.”

*607“There are two methods of curing or partly curing this high temperature defect. In small articles, heating tliem to a yellow or orange tint and then, reforging or rerollhig is an effective operation, but as this cannot be applied to large objects like armor-plates, without destroying the specially carburized layer on one face which has just been produced, another property of steel is made use of which is this. Steel which has been injured by long-continued high temperature, and them suddenly chilled, can have to a great de-

*608Now, if this be a process of annealing, and his testimony in many places refers to it as an annealing process, it must be carried on below the critical point, and this appears from the testimony of Prof. Langley to be the fact.3

“Annealing” was a process known and recognized in the prior art, and, indeed, it would seem to us that, subject to the different practices incident to the different objects to be annealed mid the particular results desired, the term “fiber” or “fibrous” as applied to metals was a term of quality rather than a description of a species of structure. Prof. Langley speaks of the grain, “or as it has been elsewhere called in this suit 'fiber.’ ” Prof. Hartshorne says:

“The term ‘fibrous’ or ‘fiber’ as applied to steel is a misnomer. No steel really has any fiber. What the term means is that the steel is in such a condition of toughness that, when the bar breaks on bending, there is an appearance as of fiber through the elongation of the particles, so that the term ‘fibrous,’ as applied to steel, merely means that the piece breaks with an appearance like wooden fiber, rather than short off like glass.”

Prof. Howe does not deny this statement, for he concedes one cannot see the fiber until the steel is broken, and this seems to be the opinion of the patentee, for he describes it as having “an appearance on fracture very similar to velvet with a silky border.” Mr. Unger, the head of the Carnegie Company’s Research Bureau, is quoted by Prof. Plowe as stating that he has never been able to detect fiber with the microscope in an unstrained specimen. And not only is this the case, but Prof. Langley concedes that the fibrous grain is not the result of this second step alone of the Schmitz process, but is in a measure the result of a subsequent step.4

Taking all the proof, we are of opinion that the second step of the patent, viz., “reheating to a miximum temperature below that permitting retroversion of the granular structure,” is a reheating below the *609critical point, and then “allowing the plate to cool slowly” is simply an annealing process.

Such being the case, the next step is also a reheating below the critical point, for it is restricted to “a temperature equal to or somewhat lower than the temperature of the second reheating.” It is contended, however, that this reheating must be above the critical point, since the claim says it must be “not so low as to exclude hardening,” but, if this be the true construction of the patent and claim, then the patentee has not fulfilled the statutory requirements of dearness, and we agree with the conclusion of the witness Hartshorne, who says:

“There is no legitimate construction of this claim which does not involve, as T understand it, the manifest absurdity of instructing the operator to heat above the critical point and to heat below the critical point at the same time, or else instructs him to heat below the critical point and then harden by sudden chilling, which is an impossibility.'’

Following the directions of the specification and observing the limitations of the claims, we are clear in holding that the temperature of the second step is at the same or a lower temperature. Such being the case, we are equally clear that it involved no invention to use these steps in armor-plate in combination with grain-refining. To our mind the reading of the patent is clear as to temperatures, and, unless uncertainty and confusion is raised by the expert testimony, none exists. The directions as to these temperatures being below the retroversion point of the granular structure are explicit, and restricted to that point, the steps are simply annealing ones. We accordingly hold the claim void as not involving invention.

As the remaining two patents, viz., No. 651,965, granted June 19, 1900, to Ehrensberger, and No. 653,413, granted July 10, 1900, which, respectively, form the subject-matter of the third and fourth bills, concern the same general subject-matter, we consider them together. They were originally applied for in one application, but were subdivided by the Patent Office, 'filie object of the patentee was to further harden the front and toughen the rear side of armor-plate, and this was done by simultaneous, differential heating of the plate, and suddenly cooling it. These are called the “differential heating patents.” The process is thus stated by the patentee:

“One side, viz., the side intended to be exposed to the impact of pro.iectiies, is heated to a temperature adapted to produce the required degree of hardness. This X call the ‘hardening’ heat, and the other or rear side is heated to a temperature below that to which the exposed side is heated. This hi Her temperature should be such as to produce toughness in the highest degree, and will be a red heat. This I call the ‘toughening’ heat. The plate is then suddenly cooled either by immersion or spraying.”

An examination of the file wrapper shows that, as originally filed in 1896, the specification did not specify the heat to which the back was to be raised, except that it was to be lower than the heat of the front. The language was that the front face was to be heated “to the temperature adapted to produce the desired suitable hardness, whilst the back or rear surface is on the contrary heated to a lower temperature; that is to say, to that temperature which has been found by experiment adapted to produce the greatest degree of toughness.” *610Under stress of rejection, persisted in by the department through a controversy that lasted four "years, the patentee in 1900 amended his specification, inter alia, by the provision we have quoted, specifying that the back was to be raised to a red heat. This he did to avoid an English patent cited against him, and in differentiating such reference he maintained the back of his plate was raised to a red heat. His language referring to the patent cited against him was:

“This patent states that the body of the gun should never exceed a black heat. Claims 1 and 2 of the present application state, however, that the rear side of the plate should be heated to a ‘lower red heat.’ Page 2, line 5, of the description, also states, ‘Heated to a lower temperature; that is to say, to that temperature which has been found by experiment adapted to produce the greatest degree of toughness.’ This shows that in applicant’s process this temperature of the rear of the plate is always a definite one; that is, one which has been proved by experiment to be the proper temperature for the particular steel. This distinguishes applicant’s process from all of the inventions ’cited. According to applicant’s claims, the low temperature is always a red heat, the temperature of which is determined according to the particular steel being treated.”

So also in the file wrapper of his other patent here involved, viz., No. 653,413, Ehrensberger made this statement in order to avoid the same reference (Kidd, British patent 3614, 1889) :

“Kidd distinctly states that the head part is heated to a bright red heat declining to almost a black at the top of the sand. If the shell is black at the top of the sand, it certainly must be somewhat cooler below the level of the sand, and consequently nothing in the nature of a toughening heat is reached by Kidd for the body of the- shell. This fact would preclude Kidd’s process from anticipating applicant’s process wherein the rear side of the plate is heated to a, good red heat."

What Ehrensberger meant by a “good red heat” is probably what is known in the art as a visible red, for such seems to have been the practice of those using the process. Thus Langley Says the back of the Carnegie plates in differential heating were 500° to 575° Centigrade, and Unger says at 550'° a plate would show at night “a very dark red.” Indeed, the various stages of red heat, as distinctive thermal conditions, were so well recognized in the steel art as to justify Prof. Langley, after referring to the stages known as “red heat just visible,” “dull red heat,” etc., to say the various values of even the different red tints was an universally recognized fact in the steel-making art. He further says:

“Until the somewhat recent Invention of accurate pyrometers, the heat treatment of steel was accomplished by reference to the color of the hot metal, and these colors are still very largely used in shop practice because to the trained eye they are fairly definite, and the lower tints are quite accurately gauged.”

[8] In view of the constrained amended language of the specification stating that the temperature of the back of the plate in Ehrensberger’s process “will be a red heat,” and the differentiations made by the applicant of the temperature of his back being a good red heat as contrasted with a black, the red he°at of the back of the plate must be read into his claims, for the principle is too clear to require citation of authority that a patentee cannot make an infringement of the *611tiling from which he differentiated his invention in order to obtain Ids patent.

[9 ] Without, therefore, discussing the grave question whether these patents disclosed a workable invention, we are satisfied the complainant has not established the fact of infringement. This fact turns on the testimony of Leonard and Ross, neither of whom had the general knowledge of the armor-plate art which Mr. Unger testifies was confined to a very narrow circle of 15 or 20 men in the United States. Their testimony was the best the complainant could get under the circumstances, as the respondent, like all other armor-plate manufacturers, carry on their operations secretly, and these witnesses were men who while engaged as workmen in respondent’s establishment were in fact under the pay of another employer, who engaged them to observe and furnish information. Who their employer was docs not appear in the proofs, so that the fact that they were not bona fide workmen is in no way a reflection on the complainant. It is only referred to to show the .character of such witnesses in connection with the statement of the court below that the evidence of infringement was “not sufficiently satisfactory either in quality or amount.” Ross, one of the witnesses, was not shown to have any special knowledge of or any practical experience whatever in armor-plate work. lie made his observations while running a power crane in the Midvale plant. His whole testimony in reference to the differential heating is as follows:

‘■Q. Yon have stated that, after the plate is machined ready for face hardening, 'the plate is put in a bed of yellow clay, the clay is fetched up all around the edge of the surface of the plate.’ Is the face of the plate covered with clay? A. No, sir.
“Q. You then state, ‘The plate is heated up, then it is fetched out and put on a spray and dashed with cold water.’ Did you ever note the color of the plate after it is ‘heated up,’ and fetched out to be sprayed? A. Yes, sir.
“Q. What is the color of the face of the plate? A. About a cherry red.
“Q. What is the color of the back or the bottom of the plate? Á. Black.”

Beyond stating that he worked both day and night, that the colors to which he swore were those which he saw after the car bottom was run out of the furnace, and that the process to which he testified was the one he saw practiced by the Midvale Company (luring the two years he operated the crane in the armor tempering department, the foregoing is his entire testimony bearing on the fact of infringement. The quoted statements were brought out by the complainant’s own questions, and they positively and without qualification establish the fact that the Midvale Company during the two years of the witness’ day and night observations kept the backs of their plates black. Not only does this testimony fail to show that Midvale used any infringing process, but it went further, and positively and unequivocally established affirmatively that it did not infringe, because, as the witness testified, the backs of the plates were black, and not red.

The only other witness was Leonard. His experience in iron and steel work was confined to puddling iron in mills and heating in open hearth furnaces. He acted as a watchman at Midvale for about two *612months, and for the ten months following worked under office instructions, heating armor-plate. His testimony on the subject was:

“Q. Please describe the face hardening process? A. The plate is then taken and put on a form the shape of the plate and bedded in mud until about an inch and a half of the plate is exposed. Then it is put into a white-hot furnace, and heated until the plate becomes a good cherry red, or about 730° Centigrade. Then it is taken out and put on the spray and sprayed from the top and bottom both at the same time until the plate is perfectly cool.
“Q. What is the plate sprayed with? A. With water, cold as ice water.
“Q. Which side of the plate is covered with mud. and which side is exposed? A. The face of the plate, with the carbonizing material, is exposed to the heat of the furnace, and the other part of the piafe is embedded in mud.
“Q. You state the plate is heated until it becomes a cherry red. Does it become a cherry red all over the plate? By that, I mean all over the surface of the plate? A. It is heated until the face of the plate becomes a cherry red all over. That is an equal heat' all over the face of the plate.
“Q. What is the color of the back of the plate, or that portion of the plate covered with mud? A. I saw them black, and sometimes showing the color pretty good, but, if the bottom of the plate was as hot as the face of the plate, that was considered a bad heat, and it had to be heated over again.”

Beyond stating on cross-examination that “the idea, and my belief was, to keep the back of that plate as cool as they possibly could,” the foregoing constitutes Leonard’s testimony. This testimony also fails to establish infringement. It shows the effort at Midvale was .to keep the back of the plate as cool as could possibly be done. This corresponds with what Ross says, who during two years always saw the backs black. The further testimony of Leonard that he saw the backs of the plates “sometimes showing the color pretty good, but, if the bottom of the plate was as hot as the face of the plate, that was considered a bad heat, and it had to be heated over again,” is confused, and the experts'havé put different constructions upon it.

But infringement is a tort, the burden of establishing which is on him who charges it. To produce confused and uncertain testimony in that regard will not suffice. Taking the testimony of Leonard and Ross as a whole, we have no doubt that it not only fails to show that in differential heating the Midvale Company did not heat the backs of their plates to a red heat, that their object was to keep them as cool as they could, and that the backs of their plates showed a black heat. This being the fact, we are of opinion the complainant has not proven infringement of the claims in issue in Ehrensberger patents, Nos. 651,-965 and 653,413.

. The decree of the court below will therefore be affirmed, with costs, but our affirmance of such decree is based on our conclusion that claim 5 of patent No. 534,178, granted to Albert Schmitz and Emil Ehrensberger, is invalid; that claim 1 of the patent No. 642,926, granted to Albert Schmitz, is invalid; that claims 1 and 3 of patent No. 651,965 and claims 1 and 3 of patent No. 653,413, both granted to Ehrensberger, are not infringed.

This conclusion is in accord with complainant's proofs. Exhibit No. 24, Senate Document No. 141. Report of Chief of Bureau of Ordinance to Sen-ale February 20, 1899, says: “The process of carburization, which is accomplished in the Harvey process by charcoal under a high and prolonged heat, is accomplished by means of a hydro-carbon gas in the case of the Krupp plates, and requires furnaces specially adapted for the purpose.”

Deference may be made to these authorities in addition to those in the record: Percy’s Metallurgy (Ed. 1864) p. 10; Bradley ¡Stoughton’s, The .Metallurgy of Iron & ¡Steel (Ed. 1S)08) p. 58; Treatment of Gun Steel by Ool. Eardley Maitland, yol. 89, p. 120, Minutes oí Proceedings of Civil Engineers, 1887; The Metallograpfaist, p. 97, 1899, article of Stead; 34 Journal of the Iron & Steel Institute, 1889, p. 371; 7 Transactions of American ¡Society of Mechanical Engineers, p. 241, 1886; Howe’s Metallurgy of ¡Steel (Ed. 1890)’ pp. 194, 196, and instances cited; Tieman’s Iron & Steel (Ed. 1910) p. 89.

Thus Prof. Langley says: “The directions "just quoted from the patent [now, if this plate thus hardened be again subjected to a glowing heat, etc.] denote that the glowing heat which is not to be so high as the previous hardening heat, and is to be about 600 degrees Centigrade, is clearly below the critical heat of hardening. It does, however, have the effect of annealing the plate, and thereby removes the hardness communicated by the previous operation by rapidly cooling from 800 degrees Centigrade.”

“Q. If an armor plate, say of the character you saw at Homestead has been properly annealed after carburization at a temperature below the critical point, is subsequently reannealed at the same or a lower temperature, what effect would such latter heat have upon the structure produced by the earlier annealing? A. In my opinion the second annealing would have the effect •of continuing and confirming, so to speak, the condition of grain existing at the termination of the first 'annealing.”