General Electric Co. v. Independent Lamp & Wire Co.

MORRIS, District Judge. The plaintiff, General Electric Company, by its bill charges Independent Lamp & Wire Company, Incorporated, the defendant, with infringement of United States letters patent No. 1,082,933, granted December 30, 1913, to plaintiff, as assignee of William B. Coolidge, for improvements in tungsten and — 825 — methods of making the same for use as filaments of incandescent electric lamps, and for other purposes. The application for the patent in issue was in part a continuation of prior applications by Coolidge, filed July 2, 1906, October 6, 1909, February 23, 1910, and August 15, 1910. [1,2] The defenses are invalidity of the patent and noninfringemeni of certain claims. The specification states: “My invention comprises a new incandescent lamp filament of drawn wire made from the metal tungsten and a process of producing the same. * * * The wire produced hy my invention has found various other applications. * * * So also the new material of which the wire consists, fully worked ductile tungsten, has in other mechanical forms a wide variety of useful applications. Further, an incidental, but valuable, new product is found at an intermediate stage of my process. * * * ” Touching the discovery the specification says: “I have discovered a process by which tungsten bodies, when prepared under certain conditions, as will be.hereinafter more fully described, can be mechanically worked, as hy hammering, swaging, rolling, and drawing, and have further discovered that, when this mechanical working is carried on while the metal is heated to temperatures within certain maximum and minimum limits, and is continued long enough, such bodies will he converted from their original crystalline character to a condition having all the characteristics of a ductile metal. In other words, the repeated hot working so changes the metal that it acquires tensile strength, and also becomes pliable and ductile at ordinary or room temperatures, and if the mechanical working is carried sufficiently far, ana under proper conditions, the tensile strength of the metal may become equal to or greater than that of the best steel. I have thus not only devised a new process by which it has become possible to work tungsten, as by hammering it or rolling it into the desired form, and by drawing it into fine, strong wire, hut I have also produced by the operation of this process a new product, viz. wrought tungsten, and, if the process be carried far enough, ductile tungsten, a material having properties and utilities different from those of any previously known substance. When it is desired to produce from this material an incandescent lamp filament, or any other body which in normal operation is to be operated at high temperatures, I use a special mode of preparation in order to minimize the tendency of the tungsten to revert to a brittle crystalline form. When this crystallization becomes excessive, the crystals may, in the case of a filament, become so large as to extend across the entire section of the filament, and thereupon the sections may move laterally upon each other and produce the condition known as ‘offsetting.’” The claims are for process and product. Typical claims are: “1. The process of producing tungsten having a fibrous structure which consists in repeatedly hot working a crystalline body of tungsten until the crystalline structure is broken down and a fibrous structure developed.” “24. A wire formed of ductile tungsten. “25. An incandescent electric lamp having a filament of drawn tungsten wire.” The history of tungsten for upwards of a century after its discovery in 1781 is brief and monotonous. It does not occur native in nature, but is found in certain ores, as wolfram and scheelite. Its ores exist in abundance in all parts of the world. The metal is isolated as a heavy steel gray to black powder, having a fusing point of about 3,200 to 3,350 degrees Centigrade. Since discovery it has been almost constantly under the inquiring eye of chemist and metallurgist. Large quantities of tungsten and its compounds have long been used in alloys, such as tungsten steel, and in chemical compounds. It was well — 826 — known that it possessed certain properties that would make its use greatly beneficial, particularity to the electrical art. As a powder, however, it was not capable of being so used as to make its desirable properties available. ' Owing to its high specific gravity, it was also thought desirable for projectiles. From the beginning of the nineteenth century efforts were made to cast it, reduce it to plates, or draw it into wires; but, with certain possible excéptions which will be hereinafter considered, it was with frequency and uniformity pronounced hard, brittle, unworkable, and nonductile. As late as 1903 Sir Robert Hadfield, a leading metallurgist, in a paper read before the British Iron & Steel Institute, said: “As far as we know, the metal tungsten, like cliromium, is not malleable. If an absolutely pure metal could be obtained, possibly this statement might have to be modified; but the purest forms which the author has been able to obtain possess hardness, brittleness, and are not ductile, either in the ordinary or heated condition.” In 1904 two Austrians, Just and Hanamaii, succeeded in making a tungsten filament, not by drawing, but by mixing the fine tungsten powder into a wet paste, with starch or sugar to act as a binder, squirting the paste into a thread, drying and baking the threads, hanging up each thread in an atmosphere of moist hydrogen, passing electric current through it, whereby the carbon in the binding material was removed by the oxidizing action of the moisture in the hydrogen, and the particles of tungsten sintered or softened and stuck together until the threads became consolidated into filaments, which, though exceedingly brittle, were usable in incandescent lamps. This was the first practical use ever made of the 'metal tungsten, pure and unalloyed. The patent for this invention was sustained in General Electric Co. v. Laco-Philips Co., 233 Fed. 96, 147 C. C. A. 166. The advantages of tungsten as an incandescent lamp filament are due chiefly to its extremely high melting point and its low vapor tension at high temperature ; that is, it may be heated well up towards its melting point, even in a vacuum, which generally helps volatilization, without giving off. vapor sufficient to cause appreciable deterioration, though the lamp be burned hundreds of hours. The Just & Hanaman filament, brittle though it was, brought about the enormous advantage of diminishing the power necessary for a given illumination to approximately one-third of the amount formerly required, and produced a whiter light. The revolution in the art caused by the Just & Hanaman invention was described by Judge Mayer in his opinion in the case above referred to. But, notwithstanding their advantages, the Just & Hanaman filaments had many shortcomings. They had to be made in short loops or hairpin shaped sections, of which several were required for a single lamp. They could not be made with sufficient precision to meet definite voltage requirements. They were exceedingly fragile. They broke in great numbers in manufacture, in shipment, and through vibration when in use. Though Just & Hanaman gave tq the world the tungsten lamp, they did not advance the art, if art there was, of working tungsten. They did not work it. They did not even discover that it was susceptible of being worked. They did not give to it any new artificial — 827 — properties or characteristics. They did not make it pliable or ductile. Chemists and metallurgists continued to say that tungsten, whether cold or hot, was neither malleable nor ductile; that it could not be drawn into wires. For instance, in “The Petty Metals, Titanium, Tungsten, Molybdenum,” by Truchot, published in 1905-06, it is said: “Like chromium, tungsten is not malleable or ductile, whether cohl or hot. When pure, it is hard and brittle.” In Roscoe & Schorlemmer’s Treatise on Chemistry, revised by Sir H. E. Roscoe, F. R. S., and Dr. A. Harden, in 1907, it was stated: “The purest forms of tungsten at present obtainable are hard and brittle, "and are not ductile, cither at ordinary temperatures or when heated.” Even as late as 1910,' in “Illumination and Photometry,” by Prof. Wickenden, of the Massachusetts Institute of Technology, this statement occurs: “The metallurgy of tungsten is very complex, and the metal, when refined, is nonduetile.” Consequently a problem remained to be solved. It was said: “The elimination of this excessive brittleness is the crux of the tungsten lamp problem.” The specific problem was how to make ductile a metal naturally and normally nonduetile. To the solution of this problem the lamp manufacturers in Europe and America bent their every energy. Coolidge solved it, and produced wrought tungsten, or “Coolidge metal,” having properties differing most radically from those of the normal or natural tungsten metal. Tungsten is nonduetile; Coolidge metal is remarkably ductile. Tungsten is absolutely brittle at ordinary temperatures ; Coolidge metal is pliable and flexible. Tungsten is fragile and easily broken; Coolidge metal is stronger than steel. The new metal is so ductile it may be drawn into wires uniform in quality miles in length; it is so flexible that it can be tied in a knot or used as a thread to sew on buttons; its tensile strength may run as higli as 600,-000 pounds per square inch, while the tensile strength of steel piano wire, the material next in strength, is only about 400,000 pounds to the square inch. The announcement of the Coolidge achievement caused much comment. The United States Geological Survey Report for 1910, “Mineral Resources of the United States,” published in 1911, says: “No important new uses for tungsten came to the notice of the Survey during the year, but suen wonderful improvements were made in the manufacture of tungsten incandescent electric lamp filaments as to make their use amount almost to a new one. The General Electric Company so developed the drawing of tungsten into fine wire that, as now made, it is about as strong as steel wire. This has made possible, not only the shipping of tungsten lamps with a. very small loss from breakage of the filament, but railroad trains and automobiles now carry tungsten incandescent electric lamps. To railroads this means a very great saving in electric power, with a consequent lessening of the weight of storage batteries used or a lengthening of their period of service; per charge: or it means that a smaller dynamo will light the trains. * * * The toughening of the filament has also made more practicable lamps of large candle power. Tungsten lamps are now used almost everywhere that electric lighting is used.” — 828 — Numerous other publications spoke of it in the highest terms. Prof. Thompson testified: “The advantage given by the ductile tungsten wire of Coolidge which might be briefly referred to as ‘Coolidge metal,’ is an avoidance of brittleness in manufacture; the filament being handled like any fine wire, wound on a frame or properly looped, with only its ends attached to the wires leading through the glass of the lamp for electric current, instead of, as in the sintered filament, in the forms used involving a number of joints, two for each loop or bent portion. Another advantage, which is of great value, is that whereas, with a sintered filament, which preceded the Coolidge wire, it was more or less of an accident whether the lamp came out with its proper voltage rating or not, .while with the Coolidge wire it is the practice to cut a definite length of wire for a given voltage and mount it in the lamp. The making-of the sintered filament lamp was like shooting at a target, with occasional bull’s-eyes. The making of the Coolidge ductile tungsten lamp was as if the target was every time reached by a dead shot in the center. * * * The whole incandescent lamp manufacture was revolutionized by the Coolidge ductile tungsten filament, without loss of those advantages which had already been secured in incandescent lighting by the sintered filament. * * * It has also given rise to a new form of incandescent lamp, known as the half-watt or gas-filled incandescent lamp, in which the filament is a tiny close coil of tungsten wire, mounted and used as an iHuminant, and surrounded by an inert gas, such as nitrogen or argon, enabling the consumption of energy for a given candle power, which had been reduced by the introduction of tungsten to about one-half to one-third of what it was with the prior carbon filament lamps, to again be cut down in about the ratio of one-half; that is, one-quarter or one-sixth of the energy for a given illumination in the case of carbon.” The defendant, too, adds its verbal tribute. In January, 1915, it advertised its lamp, which is substantially the same as that of the patent in suit, as follows: “The Life of a Lamp Lies in Its Filament. “Independent Ductile Drawn Tungsten Wire, As a filament for tungsten lamps, represents a tremendous step in the development of lamp manufacture. It gives Independent lamps a durability far greater than is obtained with the pressed tungsten filaments. The Independent, not only will resist shock and vibration, destructive to the ordinary lamp, but it furnishes a white light, more nearly approaching daylight than anything yet devised. “Pick your lamps for long, satisfactory, economical service. Get our details and prices.” And again: “The Vibration Kills Tour Lamps Unless the Filament is Ductile. “Ductile metal — metal that can be drawn into wire — cannot be brittle. In fact, Webster’s Dictionary gives ‘pliant’ as á synonym for ‘ductile.’ “Independent Ductile Drawn Tungsten Wire Stands the Racket. Of vibration because it is pliant — mot brittle. That’s why Independent lamps give longer service than the ordinary kind. Specify Independent — the lamps with the durable ductile drawn -filaments. Write for quotations.” The ductile tungsten filament has gone into general use, and has in large measure, if not entirely, supplanted all other filaments, including that of Just & Hanaman, for incandescent electric lamps. It is less expensive to produce, yet it has far greater durability and utility than — 829 — the squirted filament. It is a commercially new product of great utility, having properties and characteristics unknown to the natural and normal tungsten, or, in fact, to any other material. The defendant contends, however, as I understand its argument, that tungsten is ductile, or it could not be drawn into wires, and again that the drawn tungsten is not ductile, in that annealing makes it brittle, instead of more workable. The first contention overlooks the distinction between natural and artificial ductility, or ductility produced by mechanical strain. Such metals as nickel, silver, and copper are naturally ductile; worked under certain conditions they become very brittle and are readily broken. Yet it would not be deemed accurate to say that such metals are brittle. Again, as I see it, annealing is not a test for ductility. It was the one means known to the prior metal working art for restoring to a metal the ductility destroyed by mechanical strain. But its restoration presupposes prior ductility. Consequently the test for ductility is not annealing, but in ascertaining whether the metal is capable of being permanently drawn out or hammered thin without being fractured. The defendant also urges that the work of Coolidge was done in the metal working art, and that in this art it was well known that the application of heat and mechanical forces would produce certain results, including those of Coolidge; that all the knowledge necessary to make the Coolidge product is embodied in the adage, “Strike whilst the iron is hot.” The plaintiff, on the other hand, meets this contention of the defendant by the statement that the characteristics of tungsten are peculiar to it; that there was in reality no prior art as to tungsten, and that the metal working processes of the prior art taught that the application of mechanical forces to a metal at a temperature below its annealing point produced brittleness and hardness, not ductility, malleability, and pliability; that working hot or cold had never been known to produce these properties, at least after cooling, in a metal normally and naturally brittle. These conflicting contentions make necessary an inquiry touching the prior metal working art. In the main, T find the testimony of Prof. Zay Jefferies the most satisfactory upon this point. Metal working is performed by changing the shape of the metal by means of mechanically applied stresses, such as hammering, rolling, and wire drawing. Before the metal can be worked, it must, of course, be first obtained from the mine and separated from its ores, if it be not found native, and presented to the mechanic in a mass suitable for working. The mass so presented to the mechanic is known as an “ingot,” “slug,” “billet,” or, where the mass is small, a “button.” To produce the mass for working, the metal is invariably subjected to a temperature either above or just below its melting point and allowed to cool. When cool, the metal, whatever it may be, will be crystalline in structure. Consequently the metal as received by the mechanic is always granular, not fibrous. The number and size of grains vary according to the rate of cooling; rapid cooling producing numerous grains, or a “fine-grained” structure, and slow cooling producing fewer and larger grains, or a “coarse-grained” structure. The normal shape of grains, when free — 830 — from mechanical strain, is equiaxial; the diameter in all' directions being approximately the same. Certain metals, such as gold, silver, platinum, aluminum, iron, nickel, and copper, when composed of an aggregate of equiaxed grains, may be reshaped cold without rupture. Such metals are called “ductile.” Other metals, such as chromium, manganese, antimony, and tungsten, lack this property and are called “brittle.” When a ductile metal is worked at ordinary or room temperature, certain well-known changes occur in its internal structure and in its properties or characteristics. The individual grains are thereby deformed in about the same manner and to the same extent as the mass. Its tensile strength increases, but its malleability and ductility decrease. If the working is continued, malleability and ductility disappear, and the metal becomes brittle and unworkable. The same results follow if the working is done hot, but below the temperature at which the grains, when deformed, spontaneously reassume their equiaxial form. Reheating to a proper temperature, the metal made brittle by working restores its deformed grains to their former equiaxial condition. The restoration of the original granular structure is accompanied by a restoration of ductility and malleability, and consequently of workability. This heating process is known as “annealing.” The annealing temperature varies with the different metals. Sometimes a metal is worked above its annealing temperature. This closes the pores, and may with the use of less power make a change in the shape of the mass. If the work is finished at a temperature only slightly above the annealing point, the grain is refined and the metal made more workable at lower temperatures, for, the smaller the grain of a ductile metal, the greater its tensible strength, malleability, and ductility. The prior art also taught that hot wire drawing is the exception and not the rule. The rule and its underlying reason is stated in “General Metallurgy,” by Prof. Hofman, of the Massachusetts Institute of Technology, thus: “As the tensile strength of a metal decreases with a rise of temperature, metals are always drawn cold.” Tensile strength is, of course, an indispensable prerequisite to wire drawing. “Metallurgy” by Prof. Rhead, of the Manchester Institute of Technology, says: “In wire drawing the metal becomes hard and brittle, and required to be frequently annealed. The ductility is much less hot than cold, so that all wires are drawn cold.” The defendant cites specifically the working of zinc and molybdenum, and the Wollaston method of working platinum, as pointing the way to work tungsten. Zinc does not differ in principie from the other ductile metals, although its unusually low annealing temperature and the consequences flowing therefrom might create a different first impression. The ordinal ductile metals are less ductile when coarse-grained. Zinc, when possessing a coarse-grained structure, is brittle at room temperature, owing to the coarse grains. If the coarse grains are broken up by mechanical working above its annealing temperature, the resulting fine-grained structure is ductile at room temperature, as are — 831 — the ordinary ductile metals, when possessing a similar structure. When worked below its annealing temperature, a fibrous structure is developed, resulting in brittleness of the metal. Annealing restores its ductility. I have been unable to conclude from the record that pliable or ductile molybdenum was known to the prior art. Nor do I find that, prior to Coolidge’s efforts to work tungsten, there was any known method of working molybdenum. True, in his endeavor to attain the end sought with tungsten, Coolidge did not limit his experiments to tungsten alone. Molybdenum was included, with the hope that thereby he might acquire experience and knowledge that would be of assistance in solving the tungsten problem. In so doing he succeeded in drawing some molybdenum filaments. Coolidge’s work with molybdenum was only incidental to his tungsten work. I do not understand that such work upon molybdenum, or its result, is of any evidential value in this case, other than to show the difficulties of working tungsten. Wollaston published a paper in 1828, entitled “A Method of Making Platina Malleable.” But pure crystalline platinum is exceeded in ductility by gold and silver only. The platinum problem was not the itingsten problem. Dr. Alfred Riche, about 1857, tried the platinum processes on tungsten and failed. A manufacturer of platinum, whose skill was well known, used his every method and effort upon tungsten and failed. For more than three-quarters of a century chemists and metallurgists all over the world, with Wollaston’s work before them, were endeavoring to find a method of working tungsten, and alike they all failed. Under these conditions I attach no importance to the Wollaston method of working platinum, nor to the experiments of Dr. Uiebmann, the defendant’s expert, in working tungsten, not according to the Wollaston process, but according to the Wollaston process with Coolidge improvements. Metals normally brittle and fragile, such as chromium, manganese, and antimony, are not used commercially in the pure state. Such, in brief, was the prior metal working art. We may now examine the characteristics of tungsten and the method of working it. A mass of crystalline tungsten, however fine be the grain, is entirely brittle cold. It may be worked above its annealing temperature, which is exceedingly high; but after being so worked it is still entirely brittle cold. If the mass is worked at a high temperature, but below its annealing temperature, the grains are deformed. This hardens the metal, and makes it less workable at the immediate working temperature, but increases its workability at some lower temperature, an entirely new phenomenon.. If worked at that lower temperature, the same results follow. At each stage the grains are drawn out more and more, until a very fibrous structure is produced, which is pliable and ductile cold. If the metal is heated to its annealing temperature at any working stage, or after the working has been completed, it reverts to its crystalline condition, in which it is entirely brittle cold. The most workable condition of other metals is the least workable condition of tungsten, and vice versa. It is thus seen that the characteristics of tungsten are peculiar to it, and that the method of working tungsten is an exact reversal of the — 832 — metal working processes of the prior art, although heat and mechanical forces are used in each. Never, with the exception of tungsten, have heat and mechanical forces produced ductility in a metal. Furthermore, tungsten is made ductile by working it under its annealing temperature,, a method by which brittleness in other metals is uniformly produced. The important thing in this patent, at least so far as the process claims are -concerned,' is therefore a method of procedure, not the particular means by which the method shall be practiced; yet the specifications point out, not only the method of procedure, but also complete mechanical devices whereby the method can be put into operation, though the mechanism is not claimed by the patentee. Working tungsten was the ultimate problem, not the preliminary one. It was first necessary to produce the tungsten in a proper condition; next get it into a mass, such as an ingot, billet, or slug, suitable for working; and, lastly, to work that ingot into the desired filament or wire. Each problem presented innumerable difficulties. Coolidge’s first discovery consisted in finding that a small filament of tungsten, prepared in a particular way, could be hammered at high temperatures, and its form thereby changed. Upon becoming cool, the metal so hammered was still brittle. Experiments disclosed that the brittleness was due not to impurities, but was inherent in the nature of the metal itself. Coolidge attempted to draw a filament through wire-drawing dies. He failed. Subsequently, by heating the filament, the die through which the filament was being drawn, and the tongs by which it was being drawn, he succeeded. He succeeded in putting the filament through a second die with a draft a fraction of a thousandth of an inch less than that of the first die. The process was continued with die'after die. Eventually he discovered that the filament had lost its brittleness and was ductile even when cold. A remarkable thing had been accomplished, yet the filament so ductilized was little more than a laboratory curiosity, owing to the smallness of the original mass. The process, though it produced the new product, produced it only in small quantities and at great cost. It seemed inherently inapplicable to ingots of substantial size. Nevertheless Coolidge through years of effort and failure evolved a process by which masses of tungsten substantial in size could be converted into filaments ductile when cold. The method so evolved is the process that revolutionized the electric lamp industry, and is the preferred process of the patent. As to it the specification states: “My process divides itself into two stages: First, tile preparation' of an ingot or billet of tungsten; and, second, the working of the ingot, and I find that the more successfully the ingot is prepared the more quickly and readily it can be worked; also, by using certain precautions in working, I am able to handle ingots which have not been prepared in the best possible way.” The stages of the process are interdependent. The ingot is prepared by pressing the dry metal powder (obtained from the tungstic oxide in a manner described in the patent) in a side pressure mold under great pressure, sliding the fragile resulting stick upon a piece of metál, baking it, to give it mechanical strength, and then sintering it; that is, forming it into a homogeneous, compact, nonporous ingot by — 833 — the passage through it for a proper length of time, when surrounded by an atmosphere of hydrogen, electric current sufficient to raise it to a temperature near but below the melting point. The result is a nonporous, crystalline body .of tungsten suitable for working. If the ingot is to be converted into filaments for electric lamps, a small amount of a material, such as thoria, is added to the tungsten powder for the purpose of preventing offsetting in the filament. If the wrought tungsten is not to be used where offsetting would occur, the thoria is omitted from its composition. The ends of the sintered slug are porous, as they are not thoroughly sintered. They are unworkable and are broken off. The remainder of the sintered slug is exceedingly brittle, but is subjected with the utmost carefulness to a mechanical working process at high temperatures and within prescribed ranges for the several steps. The working is performed by swaging or rolling. Precautions are taken to prevent the metal becoming too cool while being worked. The reduction of the cross-section of the ingot in each operation is very small. The working operation is continued, and the working temperature ranges are gradually reduced, until the diameter of the rod is brought to thirty-thousandths of an inch when it is ready to enter the wire-drawing die. The drawing temperatures are high, hut below the swaging temperatures. The temperatures of the drawing operation are likewise gradually reduced, as the diameter of the wire is decreased and the fibrosity increased. With the exception of molybdenum, which differs from tungsten in many substantial respects, tungsten stands alone among the metals in its characteristics and peculiarities. It is not surprising, therefore, that chemists and metallurgists were baffled by it for more than a century. If they could not succeed in working it, it is safe to assume that a mere mechanic would fail also. Cimiotti v. Comstock (C. C.) 115 Fed. 524. In my opinion the solution of the tungsten problem involved invention. The defendant, however, cites certain publications and patents to show anticipation of the Coolidge invention. Among the most important of these is the publication of Moissan, an eminent French scientist, who, in his work, “Fe Four Electrique,” published in 1897, translated by Fenher, when referring to tungsten, said: “When it is porous, like iron, it lias the property of being welded by hammering much below its melting point.” Moissan placed in a carbon crucible a mixture of tungsten oxide and carbon aud heated it by an electric arc. The outer portion of the resulting mass contained carbon absorbed from the crucible, and was very hard and brittle. The inner portion of the mass was the porous tungsten of which Moissan speaks. Coolidge tried the Moissan process. He testified: “We were able to take from the center portions which were very porous, which could be, by hot working with a hammer, compacted to a certain extent. Upon attempting to really hot work these pieces — that is, in the sense of elongating them — they cracked all to pieces. It seemed clear to me from Moissan’s publication that all he had done was to produce a porous mass ol tungsten, which could, by hammering, bo compacted to a certain extent, and — 834 — when we, with excellent facilities, tried to go further and make some use of this material, we were unable to do so.” Prof. Jeffries says: “Moissan specifically states that the product which he hammered was porous before hammering, and consequently, in order to obtain a coherent mass of workable tungsten by hammering this porous mass, welding must take place. It is of interest in this respect to note that, under the most ideal conditions for welding of tungsten particles together, I have determined the minimum temperature for a period of 15 minutes to be not less than 2,200° Centigrade. I have also tried many times, under the most ideal conditions commercially available at the present time, to weld pieces of tungsten together by means of mechanically working the tungsten in the air atmosphere after it had been heated to an "extremely high temperature in an atmosphere of hydrogen. I have not succeeded in welding the two parts together in this manner.” Moissan’s contribution was a laboratory experiment, that has never proved of any value in practice. Apparently it taught the world nothing. It is inconceivable that Just & Planaman could have attained fame by their inferior, fragile, squirted tungsten filament, if those skilled in the metal working art had learned from Moissan years before how to work tungsten. Dr. Riebmann testified that he performed the Moissan operation and says: “I obtained a product which tallied exactly with the description given by Moissan.” But, where Moissan stopped, there Dr. Riebmann stopped. I look in vain among defendant’s exhibits for a filament drawn from the porous tungsten of Moissan. I am constrained to conclude, therefore, that the Moissan metal cannot in fact be worked, and that Moissan did not in truth discover that tungsten is susceptible of being worked. But this conclusion is not supported by inference alone, for enlarged photographs of the Moissan metal, hammered by Dr. Riebmann, show very little, if any, deformation, but do §how very large cracks. Manifestly Moissan did not teach Dr. Riebmann how to work tungsten. Again, it expressly appears from the record that, to be workable, the tungsten mass must be coherent, not porous. To convert the porous mass into a coherent mass “welding must take place.’’ Yet “under the most ideal conditions commercially available at the present time” tungsten cannot be welded. It would seem that Moissan’s hammering closed only a. few pores in the spongy mass. We have seen above what Sir Robert Hadfield in 1903 said of tungsten. Moissan failed to teach him to perform the process or make the' product covered by the Coolidge patent. Just & Hanaman and Dr. Coolidge, as well as Sir Robert Hadfield and Dr. Riebmann, were uninstructed by the Moissan publication. In fact I do not find that the disclosure of the Moissan publication was sufficient to enable any one to perform the process or make the product covered by the patent in suit. It necessarily follows that, if the Moissan publication did not instruct how to work tungsten in a practical sense, ft did not disclose that tungsten could be so worked, or that Moissan discovered that it was susceptible of being so worked. “Novelty,” says Walker in his work on Patents (section 57), “is not negatived by any prior patent or printed publication, unless the information contained therein is full — 835 — enough and precise enough to enable any person skilled in the art to which it relates, to perform the process or make the thing covered by the patent sought to be anticipated.” The patents mainly relied on by the defendant are French patent, No. 358,272, of February 7, 1906, granted to Just, Hanaman, and others, an abstract thereof by Ballois in “E’Eclairage Electrique,” and a series of Siemans and Halske patents as follows: English No. 20,-277, of 1904; German, 165,138 of 1904 and 173,134 of 1905; English No. 3174 of 1907; and Austrian Nos. 33,683 and 34,416 of 1908. The Just & Hanaman patent and the article by Ballois are similar. They state that tungsten can be drawn, but do not show how this could be accomplished. Manifestly, in view of the failure of scientific men for a century preceding, and the inappropriateness of the wire-drawing processes of the prior art for drawing tungsten, the mere statement that tungsten can be drawn is no disclosure at all. A subsequent statement of Just & Hanaman is in conflict with their statement in the patent. On January 9, 1907, they filed an argument in the United States patent office in which they said: “Tungsten, wlietlier pure or impure, whether melted or not, is so brittle as to render tiie same not drawable.” Incidentally, it may be noted that Just & Hanaman made this statement after knowledge of the Moissan publication, for they refer to him in their French patent. The disclosures of English patent, No. 20,277 of 1904, and German patent, No. 165,138 of 1905, are no more instructive than those of the Just & Hanaman French patent. Siemens and Halske by their British patent endeavored to.extend their drawing process, which had been successful with tantalum, to molybdenum, thorium, titanium, tungsten, and zirconium. The problem with tantalum had been to obtain it in a pure state. When pure, it was found to be naturally ductile. The Siemens & Halske patents now under consideration were applied for apparently with the hope that tungsten and the other metals mentioned would be found ductile under like conditions. This conclusion is supported by the fact that Siemens & Halske in their British patent, No. 5,387 of 1908, say in substance that tungsten is nonductile. Dr. Eiebmann, however, performed some experiments for the purpose of demonstrating the practicability of the disclosures of the English patent, in so far as they relate to tungsten. Dr. Eiebmann was formerly an employé of the General Electric Company, where he also worked in its laboratory “on the problem of developing the ductile tungsten filament.” tie left the employ of the plaintiff about August, 1910, when Cooliclge’s work had been substantially completed, went to Europe, where he was employed until October, 1912, in connection with the manufacture of carbon lamps and squirted filament tungsten lamps. He returned to America in October, 1912, and— “started at once to install modern niaehinery for the manufacture of tungsten lamps in the factory of the Independent Lamp & Wire Company (the defendant). and also to proceed with the manufacture of ductile drawn wire for these lamps.” — 836 — • Hence it appears that Dr. Fiebmann learned how to draw tungsten according to the plaintiff’s method. The experiments conducted in supposed conformity with the English patent were made in a factory whose machinery was designed to work tungsten as Coolidge worked it in the laboratory of the plaintiff. As said in effect by the Supreme Court in Minerals Separation, Limited, v. Butte & Superior Mining Co., 250 U. S. 336-345, 39 Sup. Ct. 496, 63 L. Ed. 1019, it is always difficult to recovfer the realities of a situation long past, such as we have here;, but it is especially difficult when extensive improvements have been made in mechanical appliances for utilizing the invention, and the disclosures of the patent have revealed properties or susceptibilities of the metal theretofore unknown. How much of the modern art did Dr. Fiebmann employ, “let us say, subconsciously,” in his experiments? “How can a court in this very practical age be convinced of the absolute accuracy of experiments post, when the alleged anticipation ante died in its infancy, and the present invention captured the commercial art almost instantaneously?” Tungsten Lamp Case, 233 Fed. 96, 104, 147 C. C. A. 166, 174. But apart from this I think the decisions of the Court of Appeals of this circuit in International Curtis M. T. Co. v. Cramp, 202 Fed. 932, 121 C. C. A. 290, and Hanifen v. Godshalk, 84 Fed. 649, 28 C. C. A. 507, show that the English and British patents now under consideration áre insufficient in disclosure. The patents and publications hereinbefore discussed are the only ones of which the effective date is prior to July 2, 1906, the date on which Coolidge filed his first United States patent application for working tungsten. As hereinbefore stated the application on which the patent in suit issued was in part a continuation of the 1906 application. In so far as the earlier application discloses any invention contained in the patent in issue, the patentee is entitled to the benefit of the filing date of the earlier application. Badische Anilin & Soda Fabrik v. Klipstein (C. C.) 125 Fed. 543; Victor Talking Machine Cases (C. C.) 140 Fed. 860; Id., 145 Fed. 350, 76 C. C. A. 180; Id. (C. C.) 177 Fed. 248. This was recognized by the Patent Office. The soundness of the Patent Office ruling is recognized in briefs of counsel for the defendant. The pertinent disclosures of the 1906 application are the necessity for a homogeneous, coherent ingot, with instructions for its preparation; the necessity of working at high temperature, and certain useful expedients, such as encasing the tungsten to be worked in a sheath of some ductile metal. If its disclosures are insufficient, so also are the same or lesser disclosures of subsequent patents. If, on the other hand, its disclosures, not its claims, are sufficient, the patent now in issue is entitled to the henefit of the date of July 2, 1906, as against subsequent patents. I deem it unnecessary to analyze each of tírese patents. I find in them no information not found in the Coolidge 1906 application, other than the statement in Austrian patent, No. 33,683 of 1908, that an alloy of tungsten and a ductile metal may be drawn, which has no value here. But the Siemens & Halske patents should not be'finally put aside without a reference to their British patent, No. 5,387 of 1908, applied — 837 — for March 10, 1908, which is a date later than that of the application supporting any Siemens & Halske patent hereinbefore considered. This patent is referred to for the reason that therein they contradict the statement of their earlier patents, and say in substance that tungsten is nonductile. This patent says: “The object of the present invention is to facilitate the manufacture of incandescent electric lamps having filaments of nonductile, difficultly fusible metals. The impossibility of making directly a filament of one of these metals suitable for use in the lamp is well known. Broadly speaking, there are two methods of making what may be called the preliminary filament. According to one method, the finely divided refractory metal, like tungsten, is made into a xiaste or ductile mass, either with aid of some agglutinant or with aid of an amalgam or alloy of low melting point, such as a cadmium bismuth alloy. This mass is then squirted fo form a filament. The preliminary filament made in this manner cannot be of any considerable length and must be handled by skilled persons. “According to the other method, preliminary filaments of any desired length can be made, and can be wound up and preserved, so that the required length can be cut off when desired; moreover, this preliminary filament can be handled and applied by persons having little experience. The method consists in combining the nonductile refractory metal with one more ductile, and as a rule more easily fusible and more volatile; the combination being effected by imbedding the nonductile metal in the more ductile metal in such a manner that each particle of the former is wholly or partly surrounded by the more ductile metal, or by incasing a rod of the nonductile inetal in the more ductile metal, or by making an alloy of the nonductile and ductile metals, so that a ductile alloy is produced. “By whichever method the preliminary filament is made, the auxiliary matter, namely, the agglutinant, or amalgam, or alloy, or the more ductile inetal, is expelled by heating the filament by means of an electric current, in order to obtain the finished filament.” Evidently Siemens & Halske should have been added to the list of those uninstructed by the Moissan statement. Heating the alloy sufficiently to drive off the auxiliary matter results in a filament of tungsten, but it is brittle, for the heating restores it to its former crystalline condition. I am satisfied that novelty of the main process and product claims is not negatived by the publications and patents cited. The British Thompson-Houston Company, the owner of British patent, No. 21,513 of 1906, based upon and substantially similar in its disclosures to the Coolidge 1906 United States application, instituted an action for infringement thereof in tlie English courts against Duram, Limited. The patent was held invalid. The reasons therefor were set out at much length in the several opinions filed in the various courts in which the case was heard. The defendant in the case at bar contends that the English decision is in effect a decision upon the patent now in issue, and that the patent now under consideration should be held invalid for the reasons assigned for the invalidity of the English patent. With these contentions 1 am unable to agree. As I understand the opinions of the English courts, the crux of the decision was that the patent there in issue was invalid, owing to the invalidity of claim 1 of the patent; the law of England being that, if one claim is invalid, the whole patent falls. Murchland v. Nicholson, [1893] 10 R. P. C. 417; Robinson on Patents, vol. 3, § 971, note 2. That claim is: — 838 — “1. The method of working tungsten, which consists in subjecting the metal in a coherent form to the action of heat while it is being operated on or manipulated.” Lord Dunedin says: “It is really on the form of the claim that the judgments of the courts below proceed.” Mr. Justice Astbury said: “This is a wide claim for working pure coherent tungsten hot. It would appear from the'language of this claim that it is not limited to any specific or particular degree of working to any defined end, such as the drawing of filament wire. * * * For the purpose of considering the .validity of this claim it is unnecessary to decide whether the patentee made and published a new and valuable invention for making wire-drawn tungsten filaments, assuming his directions to have been sufficient for this purpose. I am inclined to think he may have done so. • * * * Claim 1, however, is in no sense confined to protecting any such limited invention, if made, but claims, as I have said, ‘working pure coherent tungsten hot,’ without further limitation.” [3] The defects of the English patent, as pointed out in the several opinions, have, I think, been overcome in the patent here in suit. It was there suggested that the claim was so broad as to be analogous to claiming a principle; but, as I view the claims in the United States Coolidge patent, they are not of that character, but are well within the provisions of the patent statutes. I think Morton v. N. Y. Eye Infirmary, 5 Blatchf. 116, Fed. Cas. 9865, inapplicable. For the foregoing reasons, the main product and process claims are in my opinion valid. J. E. Baker Co. v. Kennedy Refractories Co., 253 Fed. 739, 165 C. C. A. 333; Badische Anilin & Soda Fabrik v. Kalle (C. C.) 94 Fed. 163, 173; Young v. Fernie, 4 Giff. 577; Robinson on Patents, § 101, note 3, and sections 104 and 266. See, also, Loom Co. v. Higgins, 105 U. S. 580, 591, 26 L. Ed. 1177. [4] There are few claims, however, that require further consideration. Claims 33 and 34 are: “33. The material, wrought tungsten, having a specific gravity of approximately 19, or greater, and capable of being forged and worked. “34. Wrought tungsten, a solid coherent material, characterized by the presence of crystals deformed by mechanical working.” They were intended to cover the new product found at an intermediate stage of the process. The defendant says they are too broad, vague, and indefinite. They are broad, but they claim a thing unknown to the prior art — “wrought tungsten.” It has uses of great importance. One is an electric contact mechanism in place of the expensive metals, platinum and platino-iridium. It is estimated that during the year 1916 85 per cent, of the automobiles manufactured in this country used tungsten contacts. The number of such contacts used during that year was approximately 8,000,000. For this purpose wrought tungsten lasts two to five times as long as the more expensive platinum and platino-iridium. Another use for wrought tungsten is in forming electrodes in vacuum apparatus, particularly in X-ray tubes. I am inclined to think these claims should also be sustained. [5] There are also, certain claims, referred to as “beneficial additions” claims, of which No. 14 is typical. It is: — 839 — “14. Tlio process which consists in forming a body of tungsten powder containing additional material, which will prevent coarse crystallization of the tungsten at high temperature, sintering the body, and then subjecting it to hot mechanical working.” The defendant challenges their validity on the ground that they are broad enough to include the addition of a ductile metal, such as nickel, called for by the Siemens & Flalske Austrian patent, No. 33,683 of 1908, and on the further ground that they were anticipated by Federer’s French patent, No. 371,557, published in 1907, and British patent, No. 24,179 of 1906. The ductile metal was added to the tungsten to make a workable alloy, and was driven off before the filament was used. The beneficial additions of Coolidge are included to serve a purpose while the filament is in use. They are to prevent the coarse crystallization of the tungsten at high temperature. Read in the light of the specification, I think these claims are not broad enough to include the ductile metal alloy of Siemens & Halske, and are not anticipated by their Austrian patent. Federer’s patent, however, called for the addition of oxygen compounds of rare earth metals, s^lch as thorium, zirconium, erbium, cerium, lanthanum, and the like, to the tungsten paste of the squirted filament lamp, to delay the coarse crystallization of the squirted filament. The plaintiff contends that, owing to the ever-present extreme brittleness of the squirted tungsten filament, the coarse crystallization and consequent offsetting did not become a problem of that filament; further, that it is not the mere presence of the additions in the Coolidge filament that is effective, but the extraordinary action of the thoria, when drawn out into little rods or fibers, that determines the character of the crystallization of the drawn tungsten filament, and that, this makes patentability of these claims obvious; that it is not the delay of crystallization, which in any event proceeds rapidly when the filament is heated, but that the decisive thing is the form which the crystallization of the tungsten takes, and that the correct form is due, not to thoria, but to thoria drawn out into rods of a particular shape and oriented in a particular way. I am not free from doubt that this argument, though supported by the facts, overcomes the effect of the Federer patent; yet, as I am of the opinion that the main process and product claims are valid, and as the magnitude of the matters involved in this case is an assurance that errors of this court will not remain uncorrected, I more readily resolve the doubt in favor of the patent. The coarse powder and the Battersea crucible claims do not seem to require any special consideration. They are, in my opinion, valid. [6] Infringement is not seriously questioned. The defendant began the manufacture of ductile tungsten before the Coolidge patent was issued, but with machinery installed by 'Dr. Fiebmann, who was working upon tungsten for the plaintiff, in whose laboratory Coolidge solved the tungsten problem, until about the time Coolidge’s work had been completed. It also brought to its plant from the works of the Westinghouse Famp Company, a licensee of the plaintiff, and employed, a number of persons who had there learned the Coolidge process. The defendant not only makes the Coolidge product, and by the Cool — 840 — idge process, but in so doing it follows in a very large measure-the Coolidge mechanical details. The defendant shows that it no longer uses the Battersea crucible in the manufacture of its product. Its former infringement of these claims, however, is under the circumstances sufficient to call for a decree enjoining their further infringement. Morton Trust Co. v Standard Steel Car Co., 177 Fed. 931, 101 C. C. A. 211. All the claims of the patent will be held valid and infringed. A decree in conformity herewith may be prepared and submitted.