Electric Smelting & Aluminum Co. v. Pittsburg Reduction Co.

COXE, Circuit Judge. The patent in controversy was granted February 2, 1892, to Charles S. Bradley for an improvement in processes for separating, by the electric current, aluminium from its ores or compounds. The specification states that the difficulty with the process theretofore in vogue was that it was carried on by subjecting the fused ore to the action of the electric current in a crucible placed in a heating-furnace. It is said that by this process the fused ore fluxes with the crucible and the fluorine gas liberated attacks the material of the crucible and destroys it. The main object of the invention is to prevent these disastrous results by dispensing with the external application of heat to the ore, which is accomplished by employing an electric current of greater intensity than is required to produce the electrolytic decomposition alone. In this way the ore is maintained in a state of fusion by the heat developed by the passage of the current through the melted mass. The current performs two distinct functions. — 928 — First, it keeps the ore melted by having a portion of its electrical’ •energy converted into heat, and second, it effects the desired electrolytical decomposition, by which means the heat, being produced in the ore itself, is concentrated at exactly the point where it is required to keep the ore in a state of fusion. Bradley dispenses with the crucible and uses a heap of the ore itself to constitute the vessel in which the reduction takes place, which is not destroyed as was the crucible and admits of the process being continuous, nothing being required but the charging of fresh ore as fast as the reduction goes on either from without or from the sides or walls of the heap itself. The patent contains six claims, the first three being intended to •cover the process as applied to the separation of metals from any highly refractory ores or compounds. Of this group claim one may be taken as an example. It is as follows: “(1) The process of separating or dissociating metals from their highly refractory ores or compounds, nonconductors in an unfused state, of which the ores and compounds of aluminium are a type, which consists in fusing the refractory ore or compound progressively by a source of heat concentrated directly upon it rather than by an external furnace and as it becomes fused effecting electrolysis by passing an electric current therethrough between terminals which are maintained in circuit with the fused bath, whereby the process is rendered continuous, substantially as set forth.” The last three claims are limited to the application of the process to separating aluminium from its ores or compounds. Of this group ■claim four may be taken as an example, as follows: (4) The process of separating or dissociating aluminium from its ores or compounds, consisting in fusing and maintaining the fusion and electrolytically decomposing the ore or compound by the passage of the electric •current therethrough, substantially as set forth.” The Circuit Court found that the claims were not infringed and dismissed the bill. The complainant assigns error, contending that the •court erred in placing a narrow construction upon the claims, and in holding that the defendant’s process did not infringe. The complainant also contends that the court erred in holding that the success -of defendant’s process was due to the invention of Hall and not to Bradley and insists that the court should not have followed the rulings and decision of the Circuit Court for the Northern District of Ohio in an action pending between the defendant and Cowles Electric Smelting & Aluminum Company, but should have followed the decision of the Circuit Court of Appeals for the Sixth Circuit in the suit of Dowry against the said Cowles Company. The questions determined in the Ohio case are not involved in the present controversy. Lowry v. Cowles Co. (C. C.) 68 Fed. 354, reversed on appeal 79 Fed. 331, 24 C. C. A. 616. Although both parties •quote from these opinions in aid of their present contentions we think it advisable to be guided solely by the facts appearing in the present record. The question actually decided in that case was one of title and though the court made use, tentatively, of expressions which, considered apart from the context, may be regarded as applicable here, yet it is quite obvious that there was no intention to do more than de — 929 — cide the narrow question involved. Indeed, the Circuit Court of Appeals says so explicitly, at page 630, in these words: “We are not required, to pass upon the validity of the patents involved in this suit, or any of them. There is no issue 'of that kind before us.” These opinions, and others which have been delivered during the protracted litigations over the Bradley and Hall patents, are helpful upon many of the propositions now under discussion, but they cannot be regarded as controlling to such an extent as to justify the court in dispensing with an independent investigation of the issues arising in the present controversy. The application for the patent in suit was filed February 23, 1883, which may be taken as the date of Bradley’s invention. Prior to this date it was known that metals contained in ores which are conductors could be separated therefrom by electricity, but the problem' of separating metals from nonconducting ores by this method hád not been solved. The compounds of aluminium at ordinary temperature are nonconductors. It had, therefore, been the custom to place these refractory ores in a crucible, apply external heat until the ore was melted and then pass an electric current through the melted mass. In this way a pure product was obtained in small quantities, but as the specification states, the intense heat soon destroyed the crucible and the process could not be worked upon a successful commercial basis. This controversy relates solely to the separation of aluminium from its ores; three of the claims are designed to cover the process when so limited and the remaining claims relate to highly refractory ores of which the compounds of aluminium are a type. The defendant is engaged in producing aluminium. The investigation may, therefore, be confined to this one metal, for if the patent, when so limited, be valid and if the claims cover the defendant’s process it matters not what else they cover or fail to cover. We start, then, with the undisputed fact that prior to Bradley’s invention no one had ever succeeded in separating aluminium from its compounds solely by the use of electricity, or, in other words, no one had dispensed with external heat. The reference principally relied on to anticipate or limit the claims is the description of an experiment made by Sir Humphrey Davy in 1807 and published the following year in “Philosophical Transactions of the Royal Society of London.” Davy tried several experiments on the electrization of potash rendered fluid by heat and only attained his object by employing electricity as the common agent for fusion and decomposition. A small piece of potash, which had been exposed for a few seconds to the atmosphere for the purpose of producing moisture, was placed on an insulated disc of platina, connected with the negative side of the battery of the power of 250 of 6 and 4 in a state of intense activity; and a platina wire, communicating with the positive side, was brought in contact with the upper surface of the alkali. The potash soon began to fuse at both its points of electrization. There was a violent effervescence at the upper surface; at the negative surface there was no liberation of elastic fluid; but small globules having a high metallic lustre, like quicksilver, — 930 — appeared, some of which burned with explosion and bright flame as soon as they were formed, others remained, were merely tarnished and were finally covered with a white film. Numerous experiments soon showed that these globules of sodium and potassium were the substances of which Davy was in search. Herei then, was an interesting experiment but nothing more. It was not made with the ores of aluminium, but with minute pieces of soda and potash which were rendered conductive by moisture induced by previous exposure to the atmosphere. In a work on aluminium, published by Tissier in 1858, it is stated that, in 1807, Davy undertook to decompose alumina by the battery, as he had decomposed potash and soda, “but he failed completely.” In fact, Davy himself admits that his experiments in this regard were without substantial results. His method was incapable of producing results upon a commercial scale. He had no conception of progressive feeding or of a fused bath, or regulating the strength of the electric current. The experiment was undoubtedly a brilliant one, but it can no more be regarded as an anticipation of the Bradley process than it could be regarded as an infringement if made to-day for the first time. That such a claim of infringement might be asserted is, perhaps, conceivable, but that it could be sustained is an unthinkable proposition. Davy suggested the possibility of producing metal from certain ores conductive in a fused state, and this hint, for it was hardly more, undoubtedly set the chemists and electricians thinking, just as the discovery of Franklin put the idea of the telegraph into the brain of Morse and as the discovery of Watt made possible the inventions of Stephenson and Fulton. Judged by its practical results, the contribution of Davy was not as valuable as it now appears when read in the light of subsequent achievement. For three1quarters of a century chemists and electricians all over the world, with Davy’s work before them, were endeavoring to find a method of producing aluminium commercially, and they all failed. After Davy’s experiment, no further effort of the inventive faculties was required, says the defendant. All that, a manufacturer needed to do was to operate the Davy process on a large scale. To do this no expert knowledge was required. And yet manufacturers not only, but the most learned specialists of the age, permitted this treasure of inestimable value to remain in plain view before them and would not even stoop to pick it up. The fact that Davy’s experiment was permitted to lie dormant during 76 years of intense activity in chemistry, electricity and metallurgy is almost conclusive evidence that the defendant has greatly overestimated its importance. Duvivier’s experiment of 1854 has even less bearing upon the present controversy. A small piece of disthene was exposed to the electric flame, undoubtedly an electric arc, disengaged from a carbon point about as large as a drawing pencil. It was melted at the.end of three or four minutes and the aluminium, freed from its oxygen, ■showed itself at the surface of the molten material. A small globule set on the outer edge flattened out as it cooled and, when scratched with the point of a knife, it showed silver white and in hardness it re — 931 — sembled pure silver. That this was aluminium is not shown and cannot be shown, as the only test to which it was subjected was the one above stated. The entire description is meager and uncertain and by no means as definite and satisfactory as the prior experiment of Davy. The Siemens British patent of 1879 also describes the use of an electric arc for the fusion of metals, but electrolysis cannot be so accomplished. The United States patent to Ball and Guest, of January, 1881, is for an improvement in “electrical carbonizing apparatus.” It does not relate to the separation of aluminium, or any other metal, from its ores by electricity. Faure’s French patent, of 1880, describes an invention the object of which is “the manufacture of metallic sodium, of the cyanides, by fixing atmospheric nitrogen, and in general the treatment, at- high temperature, of alkaline salts or metals.” He employs electric arcs and external heat to produce high temperatures, and secures the sodium by chemical reaction. The British patent to Lane Fox, of 1878, is for “improvements in the application of electricity to lighting and heating purposes,” and discloses nothing more than the Ball and Guest patent, supra. It has no application to electrolyzing refractory ores. The article printed in the “Transactions of the American Institute of Mining Engineers,” in May, 1882, describes the process of some unknown inventor as it was explained to the Institute by Prof. Howe. Assuming this to be a “publication” within the meaning of the law it is too indeterminate to be of value as an anticipation. Prof. Howe did not pretend to have any personal knowledge of the facts and merely gave a brief statement of what had been told him. That he had reference to an arc process is evident from the following language : “A voltaic arc is then thrown across from another electrode against the carbon crucible,” which is described as the cathode. It seems to be conceded on all hands that the use of an arc constitutes “a radical and fatal departure” from the process involved in this controversy. The person who actually practiced the method of pro-< ducing aluminium described by Prof. Howe may have accomplished something of practical value and he may not. No one knows. Had he done so it is, perhaps, fairly inferable that something more would have been heard of it in the art of electrolysis. The name, at least, of so eminent an inventor would not have been permitted to remain long in obscurity. It might have taken time and persuasion but, eventually, his reluctance to being enrolled among the immortals would have been overcome. The court understands that the foregoing are all the references relied on by the defendant to anticipate the Bradley claims. This supposition may be inaccurate because, in the multitude of exhibits and maze of contradictions with which this ponderous record abounds, perfect accuracy is well-nigh impossible. It is thought that they describe the only instances where, in the prior art, there was an attempt to dispense with external heating and to utilize the electric current for the double purpose of fusion and electrolysis. That they do not anticipate, or materially restrict, the Bradley patent seems self-evident. — 932 — Extended discussion on this point is rendéred unnecessary for the reason that failure to prove anticipation is found by the Circuit Court and is hardly disputed by the defendant. The principal expert for the defendant, Dr. Chandler, whose reputation for learning and ability is well known to the courts, although of the opinion that slight modifications of the previous methods would produce the Bradley process, nevertheless admits frankly: “I do not recall any one process which, when applied to the ore of aluminium, would without any modification whatever have produced aluminium, in which process both the fusion and the electrolysis would have been accomplished by the electric current.” Not only did the electricians of the earlier art fail to produce aluminium by electricity alone, but the wrecks which strew the pathway, which Davy pointed out nearly a century ago, offer mute but impressive proof of the genius of the man who first surmounted its many obstacles and reached the destination in safety. Indeed, after numerous abortive attempts and repeated failures the electrical world seemed to have settled down into the belief that aluminium could not be produced by the sole agency of electricity. Accordingly the effort of inventors was directed to the perfection of processes in which external heat was employed to melt the ore and keep it in a fused state. The record abounds in such instances. Patent after patent is introduced claiming new methods of separating aluminium from its ores, but in every instance external fire is used to fuse the bath and maintain it in a fused condition. Many of these inventions were long after the introduction of dynamos and they continued to be made and practiced for several years after the Bradley invention. Indeed, so strongly was the inventive trend towards the employment of external heat that even the defendant’s inventor, Hall, could not be induced to dispense with its use until 1889. When the defendant’s works were’ started at Pittsburgh, in December, 1888, the pots were built to be externally heated and they were so heated for some time thereafter. The Hall patent of April 2, 1889, which was applied for July 9, 1886, three years after the Bradley application, was for improvements in the “process of reducing aluminium by electrolysis.” In this patent externally, heated crucibles are shown in the drawing and described in the specification. Since the Bradley invention aluminium, which formerly was regarded as one of the precious metals, has become as common as copper and brass and its price has been reduced from $15 per pound to 25 cents per pound. We do not intend to intimate that this marvelous change was due solely to Bradley’s invention, but simply, at this time, to emphasize the fact that it took place after Bradley’s invention. If he has done nothing to produce this result he is, of course, entitled to no consideration whatever, but if he has contributed something he is entitled to protection to the extent of that contribution, be it much or little. We have proceeded thus far to the conclusion that we are dealing with a'patent which discloses a meritorious process for producing aluminium in large quantities, the essential features of which are, first — 933 — dispensing with external heat, and, second, the use of the same electric current to produce and maintain fusion and electrolyze the ores of aluminium. We are unable to discover anything in the prior art describing this process or anything closely approximating thereto. The patent is, therefore, not anticipated and its claims are entitled to a liberal construction. The judge of the Circuit Court, after careful and painstaking research, reached the conclusion that Bradley had made a valuable invention, but he failed to grant relief to the complainant upon the' theory that the process which the defendant uses was an entirely separate invention, neither dependent upon nor subsidiary to the invention of Bradley. In this we think there was error. Hall’s achievement should be considered in the light of an improvement upon Bradley’s fundamental discovery. There can be little doubt that the defendant’s process is a valuable one and that to it is largely due the cheap aluminium of the present day. There is not the least disposition to detract from the merits of Hall or minimize his contribution to the art. Indeed, it may be conceded that, if the novel features so introduced be secured by a valid patent, he can hold the monopoly against all, Bradley included. This concession does not permit him, however, to appropriate the broad invention. He does not acquire the right to use the Bradley process simply because he has improved that process. He is entitled to enjoy what is his, but in so- doing he cannot appropriate the property of another. The record discloses nothing unusual in this regard. It is rarely that an invention develops ultimate perfection in the hands of the inventor. The test of actual use discovers defects to be remedied and suggests improvements to be made. If the inventor produces a new and useful result he does not lose his reward because he, or some one else, subsequently renders it more useful. This proposition may be made plain by an analogy taken from a kindred art. It is not perfect; no analogy is, but it will serve as an illustration. Charles F. Brush was the de jure inventor in the United States of a secondary battery having its electrodes mechanically coated with active material, as distinguished'from the electrode coated by the slow and expensive process of electrical dis-^ integration discovered by Gaston Planté. The electrodes first used by Brush were crude and incapable of commercial 'work. The lead powder was held in place by blotting paper tied by a string, and the battery succeeded, after charging, in developing a current of only sufficient strength to ring a call bell, but the principle was thus established which has since been utilized to propel heavy vans and railway carriages. Although Brush had made the broad invention Fauré made the important discovery that the active material could be placed on the supports in the form of a paste, paint or cement and to this improvement the commercial efficacy of the invention was due. Although the Brush patents were subjected to fierce attack during their entire existence it was- never thought that they could be invalidated or ignored because Faure had discovered a method of carrying out the invention which was far superior to anything discovered by Brush. No one could use Faure’s improvement without infringing the broad patent of Brush and no one could use the improvement without in — 934 — fringing the patent of Faüre. When worked together success was assured. We have, then, a valid patent with claims which, on their face, clearly cover the infringing process and yet the principal defense is noninfringement; the contention being that when these claims are construed in the light of the description and the prior art there is, no infringement. Considerable time has been devoted, in the court below and in the briefs, to a consideration of the patent to Hall, under which the defendant is said to- operate. It is thought this discussion is irrelevant, for the reason that the patent was not granted until 1889 and does not disclose the process which the defendant uses and of which the complainant complains. ' The process patented to Hall adopts external heat to produce fusion, the specification showing and describing an iron or steel carbon-lined crucible which is “placed in a suitable furnace, B, and subjected to a sufficient heat to fuse the materials placed therein.” It was only when the defendant abandoned the “furnace, B,” and adopted the Bradley method of fusing by means of the electric current that the charge of infringement was made. It seems evident that the defendant may practice the invention of the Hall patent with perfect impunity so far as the Bradley patent is concerned. Neither is it important to determine whether Hall knew of Bradley’s invention when he made the discovery which induced him to dispense with his melting pots. It is enough that it was at least three years after Bradley’s invention. Whether Hall was an independent inventor or appropriated Bradley’s idea, is utterly immaterial. What the defendant does is this: It uses a series of metal pots, the sides and bottom of each being lined with carbon, connected in series with a direct current generator, each pot holding about 450 pounds of molten bath material. The bottom- carbon lining is the cathode and a group of carbon cylinders,, three inches in diameter, suitably suspended on copper stems and connected with the positive pole of the dynamo is the anode. Suitable conductors also extend from the metal shell of the pot to the negative pole of the dynamo. The bath material consists of a compound of fluoride of aluminium, .fluoride of sodium and some fluoride of calcium. Alumina is added when the bath is in a molten condition and it is asserted that it dissolves freely upon being stirred in the bath. The electric current is sent through the solution by raising slightly the carbon rods and the alumina is decomposed, the metal going to the negative electrode at the bottom of the pot. The oxygen goes to the anode and escapes in the form of carbonic oxide gas. As the alumina is decomposed the bath is charged with fresh quantities, thus making the process continuous. No external fire is used. When a new pot is started the method usually adopted is to ladle into it the liquid material from an old pot, but sometimes a new pot is started from cold materials in which case the anodes are short-circuited for several hours and the current brings their ends and the adjacent parts of the lining to a red heat,the double fluoride is piled around the anodes and left for several hours until the compound is melted by the heat conducted from the lining. The alumina is then added and the process goes on as before. — 935 — To a layman this seems impressively similar to the description in the Bradley patent, but it is not surprising that skilled scientists, familiar with every detail of the art, have been able to point out discrepancies. That differences exist cannot be denied; that they are material is strenuously denied. To use an expression more familiar to lawyers than to electricians, the complainant contends that there has been a failure to “distinguish on principle” the defendant’s process from the process of the patent. Speaking generally, it is thought that most of the points at variance relied on can be traced directly to the improvement introduced by Hall, namely, the use of cryolite as a solvent for alumina. Were it not for this change it is hardly probable that infringement would be denied. But the change of materials does not create a new process but a new way of working the old process. The complainant’s position regarding the Hall process, as used by the defendant, is sententiously stated in one of the briefs as follows: “We contend that the process actually practiced by defendants is that of Hall minus the impracticable external heating feature shown by Hall, and plus the desirable and eminently successful internal heating feature of Bradley.” Hall starts by fusing cryolite and maintaining fusion by means of the electric current; so does Bradley; at least in the example given in the patent cryolite is the ore mentioned. If no other ore were added by Hall the processes up to this point would be identical. But Hall found that alumina, which is just as much an ore of aluminium as is cryolite, dissolves readily with cryolite as a solvent or flux, and he was thus enabled to produce a more efficient and cheaper electrolyte. The- Bradley process is not confined to cryolite or alumina; it relates to all ores or compounds of aluminium and all other refractory ores of a like type. Cryolite is mentioned as an illustration in the specification, but it might as well have mentioned alumina or any other similar ore. That there is nothing in the patent, or out of the patent, requiring the limitation of the claims to cryolite seems too plain for debate. The process may be used with cryolite alone or alumina alone or with both together, whether applied synchronically or successively. In either case the essential features of the process are appropriated. Again, it is argued that the claims are limited to an electric current twice as strong as that formerly employed when external heat was used and that defendant does not use such a current, and therefore, does not infringe. We are not satisfied that this proposition has been established. The patentee says: “In order to accomplish this object, I employ an electric current of greater strength or intensity than what would be required to produce the electrolytic decomposition alone.” And, again: “I employ, as I have already stated, an electric current sufficiently powerful not only to affect the electrolytic decomposition of the ore treated, but also to develop by its passage the heat required to keep the ore fused.” The meaning of this is obvious. A current must be employed sufficiently powerful to do the work in hand and, as more work is required, — 936 — the current must be of greater strength than that used when electrolysis alone was required of it. No amount of scientific theory can overthrow the plain fact that when additional work is required additional power must be provided. If one horse is to carry the load of two he must be stronger than either of those whose place he takes. He need not necessarily be “twice as strong,” although such a standard of comparison would be a wise one to follow. After a few trials his owner will know what changes to make in harness and load in order to economize “horse power.” 'So a current which is to fuse, maintain fusion and electrolyze must be of greater power than one whose sole vocation is to electrolyze. It has two additional burdens laid upon it which require expenditure of energy, relieved of these there is more strength for electrolysis. The patentee says further: “I have found that by using an electric current about twice as strong as would be employed to perform a given amount of electrolytic work in the ordinary way in externally heated crucibles, I am enabled to keep the ore fused according to my invention without the application of any external heat whatever.” It.must be remembered that he is here illustrating his invention “as applied in one particular case to the extraction of aluminium from its ore cryolite,” by the simple and embryonic apparatus shown in the drawings. There is no attempt to state a hard and fast rule applicable to all cases alike without reference to the size of the containing vessel, the quantity and character of the ore, and the amount of loss produced by conduction and radiation. The patentee, as required to do by statute, is simply giving those familiar with the art the information, which he has found to be of value, that, in the circumstances stated by him, the current should be not twice as strong, but “about” twice as strong as in the old method. He is not attempting to inform the skilled electrician what current will be required if. some other ore, or combination of ores, be used or if the process be employed with pots of much larger capacity, upon an immense scale and under different conditions. There is not a word in the claims limiting them to a current double the old capacity, and we are unable to perceive any reason why the claims should be eviscerated by importing into them the statements of a formula intended only as an illustration.. The patentee undoubtedly does say that, as compared with the old method, a more powerful current is employed, but he does not say that this result can only be produced by increasing the voltage. It may be accomplished by decreasing the resistance of the bath, and this is apparently what the defendant has done. The general manager of the defendant says: “We have never made at any one time a radical change in voltage. We have, however, as we improved our practice, slowly lowered the resistance, enabling us to get the required current density with less voltage.” Even if it be admitted that defendant has not raised the voltage, infringement cannot be avoided for the reason that the same result has been reached by increasing the size of the bath and of the electrodes and in this way decreasing the resistance. In one case the — 937 — voltage is increased and in the other the resistance is decreased, both embody the essence of the invention. Again, it is said that the patent provides that the heat must be generated in the fused ore itself and that the claims can be evaded by any one possessing sufficient intelligence to generate a portion of the heat elsewhere. It is admitted that in the defendant’s process some heat is generated in the fused ore and that in the Bradley process some heat is generated by the resistance of the electrodes. The fact that the defendant has made mechanical changes in tjje containing vessels and thus gets a greater proportion of heat from the electrodes than from the fused ore is a mere incident of the new construction and immaterial upon the question of infringement. A patented process cannot be appropriated because the infringer practices it with new, enlarged and improved apparatus. The suggestion that the defendant does not use the Bradley process because it does not regulate the' strength of the current by raising and lowering the e. m. f. of the generator, as provided in the specification, applies only to the sixth claim where the regulation of the current is made an element. The proposition is, however, untenable from any point of view, for the reason that defendant does regulate its current, not, it is true, in the precise manner described in the patent, but in a manner clearly its equivalent. The argument based upon the alleged distinction between “fusion” and “solution” is supported by considerations too technical and refined for practical adoption. The proof leaves no doubt as to what the defendant actually does and it seems to us a matter of no moment whether the reduction of the alumina to a fluid state is described as “fusion,” “solution” or “fluxing.” If the defendant had added fresh quantities of cryolite to the bath it would have followed the exact formula of the patent and the liquefaction would have been properly denominated as “fusion” or “solution.” The fact that the “fresh material,” or “fresh quantities of the ore or compound,” happened to be alumina instead of cryolite does not,‘in the eye of the patent law, change the nature of the process, even though one may be “fused” and the other “dissolved.” Various other limitations upon the claims are urged by which the defendant seeks to avoid infringement. They are of the same general nature and proceed upon the same initial fallacy, namely, that in a generic process patent every phenomenon observed during operation and every minute detail described must be read into the claims and that the least departure from the claims as so construed avoids infringement. Neither position is tenable. In a patent like Bradley’s the claims should be as broad as the invention and, even if unnecessary and unreasonable limitations are incorporated in the claims, the court should interpret them liberally and not permit a defendant to escape who reaches the same result by analogous means, though he may employ additional elements and improve mechanical appliances. In Ventilating Co. v. Fuller & Warren Co., 57 Fed. 626, 6 C. C. A. 481, the court says: . “The actual invention, if in conformity with the language of the claims, ■should control in the construction of patents. A strict construction should — 938 — not be resorted to if it becomes a limitation upon the actual Invention, unless such construction is required by the claim, it being understood that the construction should not go beyond and enlarge the limitations of the claim.” In Tilghman v. Proctor, 102 U. S. 707, 26 L. Ed. 279, the court says, at page 733, 102 U. S., 26 L. Ed. 279: “It is probably true, as contended for by defendants, that by the use of a small portion of lime, the process can be performed with less heat than if none is used. It may be an improvement to use the lime for that purpose; but the process remains substantially the same. The patent cannot be evaded in that way.” It is asserted that the Bradley process is not operative. Having found that the defendant is using the process and it appearing that the annual output of its works is now over seven million pounds, it seems unnecessary to enter upon an extended discussion of this proposition. There is, however, ample proof that the patented process when practiced experimentally produced aluminium and there is also proof that practically the same process was commercially operated for short periods both in this country and in Europe. The owners of the patent have not attempted to operate under it of late years, but the reasons are obvious. Some of the more important of these reasons are as follows: The protracted litigation over the title, the suit based ■upon the Hall patent, the present suit and the impossibility of commercial competition with the defendant without using the so-called Hall improvement or some other improvement equally cheap and effective. The cryolite alumina electrolyte cannot be used so long as the decision sustaining the validity of the Hall patent remains undisturbed. To attempt a discussion of all the questions mooted in the briefs would extend this decision beyond all reasonable length and would' serve no beneficial purpose. Even were it’possible to do so it is surely unnecessary to follow all of the excursions of the experts into the occult realms of electro-chemical science. Some of these trails seem to vanish into thin air, others are-lost in a desert of technicalities and of others still it is true that he who attempts to travel them is quite likely to find himself wandering aimlessly “through caverns measureless to-man.” Although the appeal included both patents the argument has been confined wholly to No. 468,148. It follows that the decree, in so far as it relates to letters patent No. 464,933, must be affirmed with the costs of this appeal, and in so far as-it relates to No. 468,148 the decree is reversed with the costs of this-appeal and the cause is remanded to the Circuit Court with instructions-to enter a decree in favor of the complainant for an injunction and an accounting, with costs.