International Nickel Company v. Ford Motor Company, 166 F. Supp. 551 (S.D.N.Y. 1958)
September 26, 1958
The INTERNATIONAL NICKEL COMPANY, Inc., Plaintiff,
FORD MOTOR COMPANY and Caswell Motor Company, Inc., Defendants.
United States District Court S. D. New York.
*552 Charles H. Walker, Sullivan & Cromwell, by George C. Sharp, John F. Dooling, Jr., and Anthony William Deller, New York City, John N. Cooper, Rynn Berry, Henry J. Zafian, Stuart A. White, Donald E. Degling and David W. Plant, New York City, of counsel, for plaintiff.
Kenyon & Kenyon, by Theodore S. Kenyon, Ralph L. Chappell, John A. Reilly, James H. Callahan, Richard A. Huettner, New York City, Hughes, Hubbard, Blair & Reed, by L. Homer Surbeck, Otis Pratt Pearsall, Howard A. Levine, New York City, Robert G. Harris, Dearborn, Mich., of counsel, for defendants.
IRVING R. KAUFMAN, District Judge.
International Nickel Co., Inc. (hereinafter INCO), plaintiff in this action, filed a complaint on August 14, 1952 alleging infringement by the defendants of 15 claims of its product patent for cast ferrous alloy, No. 2485760 issued October 25, 1949. By request and consent of the parties the case was referred to the Honorable Simon H. Rifkind as Special Master to take evidence and report his findings of fact and conclusions of law. On February 11, 1958, the Master filed his report holding all of the contested claims valid and infringed. Pursuant to motions for confirmation and modification, this report is now before the Court for review.
The patent in issue covers a cast ferrous alloy, variously referred to as nodular iron, ductile iron, spheroidal iron and S. G. iron. For purposes of consistency I shall endeavor to employ the same terminology used by the Master and will hereafter refer to the patented product as nodular iron.
Defendant Ford Motor Company is charged with wilful and deliberate infringement by manufacture, use and sale of nodular iron crankshafts.
Defendant Caswell Motor Company, Inc., a Ford dealer in New York City, is accused of having infringed by using and selling Ford cars equipped with nodular iron crankshafts.
Defendants' answer denied infringement and claimed that the patent was invalid in form and substance, anticipated *553 by the prior art and unenforceable because of alleged misuse. Defendants also interposed a counterclaim seeking a declaratory judgment of invalidity and noninfringement.
In determining the issues thus presented, the Master conducted exhaustive hearings, consuming some 27 trial days in which some 21 witnesses were heard, 750 exhibits were received and 2,400 printed pages of transcript were recorded. The Master also observed the product in manufacture in a licensed foundry. The Report of the Master reflects a careful and thorough examination and analysis. I take this opportunity to commend the Master for his scholarly treatment of a difficult subject.
It should be emphasized at the outset that the factual findings of the Master come to me armored with a strong presumption of validity. Morris Plan Industrial Bank v. Henderson, 2 Cir., 1942, 131 F.2d 975; Helene Curtis Industries v. Sales Affiliates, Inc., D.C. S.D.N.Y.1954, 121 F. Supp. 490, 494-495, affirmed 2 Cir., 233 F.2d 148, certiorari denied 1956, 352 U.S. 879, 77 S. Ct. 101, 1 L. Ed. 2d 80. Under Rule 53(e) (2), Federal Rules of Civil Procedure, 28 U.S. C., I must accept these findings unless they are clearly erroneous. As I pointed out in Sales Affiliates, the same principles apply to a master's factual findings as to those of a district court on appeal. United States v. Village of Highland Falls, 2 Cir., 154 F.2d 224, 227, certiorari denied Volkringer v. United States, 1946, 329 U.S. 720, 67 S. Ct. 54, 91 L. Ed. 624.
The requirement of Rule 53(e) (2) is especially appropriate in highly technical patent cases such as this. See Graver Tank & Mfg. Co. v. Linde Air Products Co., 1949, 336 U.S. 271, 274-275, 69 S. Ct. 535, 93 L. Ed. 672; University of Ill. Foundation v. Block Drug Co., 7 Cir., 241 F.2d 6, 10, certiorari denied 1957, 354 U.S. 922, 77 S. Ct. 1382, 1 L. Ed. 2d 1437. Here the Master's appreciation of the metallurgical subtleties involved was gained through actual observation of foundry procedures and personal microscopic study of the internal structures of metals, as well as through evaluation of voluminous oral testimony and documents. This is the antithesis of the case where the precise nature of the processes involved can be gleaned from the record. See Plax Corp. v. Precision Extruders, 3 Cir., 1957, 239 F.2d 792. Here we are dealing with findings stemming from examination of real evidence, views of processes in operation and above all from the careful evaluation of testimony, based on actual observation of the witnesses, with its requisite determination of credibility.
Numerous issues are raised by the parties' exceptions to the findings of the Master. They reach into every facet of the Master's detailed analysis. In my ensuing discussion I shall endeavor to treat the more important of these objections.
II. State of Prior Art
Ferrous casting, as distinguished from forging, machining or welding, has long been employed by the art in the production of a wide variety of articles. However, the trade encountered certain limitations concomitant with its use. These may be best exemplified by setting out the nature and character of the principal cast ferrous products, such as cast steel, cast iron and malleable iron, employed for such purposes.
Cast steel is an iron alloy with a minimum amount of carbon (less than 1.7%) and represents the optimum that can be obtained from ordinary ferrous materials in cast form from the standpoint of such properties as tensile strength, yield strength, elongation, etc. However, because of its low carbon content, cast steel has a high melting point and its production entails an elaborate and expensive process with a high capital investment in *554 specially designed furnaces. Accordingly it is a costly product to produce.
Cast iron is an alloy or iron having a carbon content in excess of 1.7%. Gray iron which is the most popular form of cast iron has a considerably lower melting temperature than cast steel and combines the advantages of lower cost and simplicity of production. Its major shortcoming is that the continuity of its matrix is disrupted by myriad graphite flakes. These flakes reduce the strength and impair the ductility and shock resistance of the cast.
Malleable iron, which has a carbon content between that of cast iron and cast steel, is a useful compromise between the two extremes. The disruptive effects of graphite flakes are reduced through a lengthy heating and annealing process which causes the carbon to occur in the form of compacted particles known as "temper carbon." This procedure is time consuming, expensive and generally feasible only with fairly thin castings. For a good many years the industry has been desirous of achieving the advantages of malleable iron without its cost, thickness limitations, and other disadvantages.
III. The Patent
A. Its Nature
The patent in question has allegedly achieved this goal. It provides for the addition of magnesium to molten iron so as to produce a small quantity of retained magnesium in the iron in its "as cast" state. This retained magnesium causes the graphite (crystallized form of carbon) to occur in spheroidal rather than flake form thereby producing a product with vastly improved physical properties. The minimal amounts of retained magnesium necessary to induce the desired effect are referred to in the patent both as "small but effective" and "0.04%."
According to the Master, the patent taught a means of vastly improving cast iron properties which the art had sought for many years. The use of magnesium as an additive to molten iron was itself a daring accomplishment. For years the literature on the subject regarded this as a hazardous procedure on the assumption that adding iron to magnesium would induce explosive reaction. However, by carefully regulating their process, INCO was able to limit the reaction and to utilize the changes induced by retained magnesium. The resultant alloys were subjected to critical analysis by INCO metallurgists. They learned that small quantities of retained magnesium produced shortening, thickening and curving of the graphite flakes with commensurate improvements in physical properties. Further analysis disclosed a strong correlation between the number of spheroids and amounts of magnesium retained. It was found, however, that a point was reached after which additional quantities of retained magnesium induced a reappearance of flaked graphite. It was on the basis of these studies that the patentees undertook to charter the outside limits of retained magnesium within which their successful results might be duplicated.
INCO filed its application for a patent on November 21st, 1947. Since then the industry has heralded INCO's contribution to the art as a broad, new pioneering discovery. Nodular iron as so produced had many advantageslow cost, good castability, strength, toughness, heat and *555 corrosion resistance, stiffness, wear resistance, machinability, and ability to damp vibrational noise. As might be expected, the patent was strenuously exploited by the industry, and adopted for use in the performance of a wide variety of tasks. The Master found that INCO's discovery represented a significant advance in the art and was of great importance to the industry. At the time of the hearings before the Master, there were 113 active licensees in the United States and 236 in foreign countries. From 1949 to 1955 the annual world-wide production of nodular iron had increased at an accelerating rate. INCO, recognizing the valuable asset it possessed in nodular iron, has not stood still in the face of its growing trade acceptance. From 1949 to 1955, INCO expended more than ¾ of a million dollars to promote its licensing program and license production of nodular iron. It is against this background of commercial success that the Master found the patent to embody a pioneering discovery of the highest dimensions which, unless voided for other reasons, achieved the dignity of patentability.
B. "Small but Effective" Claims
In construing the 7 claims which speak of a "small but effective" minimum quantity of retained magnesium the Master noted that the specification contained language of seeming limitation. But finding language to the contrary also interspersed throughout the patent the Master refused to conclude that these claims were limited to irons containing 0.04% retained magnesium. What the Master found of particular significance was that the amount of added magnesium necessary to induce the formation of spheroids was dependent on the amount of impurities encountered in the particular iron used. Sulfur and oxygen, for instance, tend to counteract and restrict the spheroid-inducing effect of magnesium and when such impurities are present, added amounts of magnesium may be needed to neutralize the inhibiting effects of such alien agents. The patent itself suggests such flexibility in the quantity of magnesium required to be added to the melt.
"If any undesirable elements which tend to combine with and/or counteract the effect of magnesium, for example, sulfur, etc. (including *556 oxygen, if any be present in the molten cast iron bath as is believed possible by some metallurgists), are present in the bath or are subsequently added, the amount of magnesium introduced should be increased by the amount required to counteract the effect of the presence of these elements or impurities by removing the elements or by otherwise overcoming their effects. Sulfur is the magnesium-counteracting element which is most likely to be present. It has been found that all the magnesium-containing agents which produced the spheroidal form of uncombined carbon obtained by the invention also introduced the magnesium in a form which combined with sulfur present in the bath with the result that the sulfur content was reduced to about 0.010% to 0.015%, e. g. 0.012%. The results also have clearly indicated that only the unconsumed excess of magnesium is then available to perform its function of controlling the form of the uncombined carbon and providing the required retained magnesium content" (Col. 10, line 59; col. 11, line 8).
The Master next inquired into the chemistry underlying the neutralizing effect magnesium had on the impurities present in the melt. He found that the magnesium necessary to overcome the weakening effect of sulfur and oxygen is not completely dissipated or consumed in the process. Rather, some of this magnesium is retained in the cast and appears in combination with sulfur and/or oxygen. However, the magnesium appearing in this form, though comprising part of the retained magnesium referred to in the patent, is ineffective in producing spheroidal formation. After examining the experts called by both parties it was the Master's considered judgment that this fact, i. e., some of the magnesium needed to counteract the effect of impurities appears in retained, albeit ineffective form, would be understood to those skilled in the art.
What the Master found, then, is that the retained magnesium referred to in the patent consists of two component partsone part which is effective in the production of spheroids, and another part which has been rendered ineffective through its combination with impurities. Consequently, though the effective retained magnesium which controls the occurrence of spheroids is constant, the ineffective retained magnesium varies directly with the amount of impurities present. It follows that the total retained magnesium can be manipulated up and down depending on the amount of magnesium necessary to combine with impurities. The Master concluded that the "small but effective" claims must be read against this background, that in view of the impossibility of predicting the amount of impurities to be encountered in each and every instance, these claims could not be limited to 0.04% as suggested by some of the language in the specification and that in fact the patent would be read by those skilled in the art as teaching that by purifying their iron they could successfully use less than 0.04% retained magnesium to obtain the results achieved in the patent.
Ford vigorously attacks this conclusion. It argues that from the instruction in the patent that the addition of more magnesium is necessary to remove oxygen and sulfur impurities, it does not follow that less magnesium is necessary if the amount of impurities is less than normal, and that in any event there is no suggestion in the patent that the amount of retained magnesium is a variable depending on the amount of the impurities. Ford urges that the quantity of magnesium needed to obtain the required amount of retained magnesium is in addition to any amounts needed to counteract or combine with impurities. *557 Thus, if 0.02% retained magnesium is desired and 0.01% of magnesium is needed to overcome the effect of the impurities, 0.03% of magnesium should be added. But, according to Ford, variation in the amount of magnesium retained is not taught by the patent and in the above illustration only 0.02% of magnesium should be considered as falling within the patent's definition of retained magnesium.
But to say that the patent teaches that more or less magnesium must be added depending on the amount of impurities but that in no event can retained magnesium fall below 0.04% assumes, contrary to the Master's express finding, that the magnesium needed to counteract impurities would be completely consumed. I will not upset the Master's findings on this point. Certainly the patent is susceptible to the construction placed upon it, though the language is far from unequivocal. In this connection the Master's finding that those skilled in the art would read the patent as prescribing an area of flexibility depending upon the amount of impurities involved is most compelling. Wood v. Underhill, 1847, 5 How. 1, 46 U.S. 1, 12 L. Ed. 23; Lever Bros. v. Procter & Gamble Mfg. Co., 4 Cir., 1943, 139 F.2d 633; American Stainless Steel Co. v. Ludlum Steel Co., 2 Cir., 1923, 290 F. 103.
Indeed, if the patentees had intended to limit all their claims to iron containing specified minimum amounts of retained magnesium one might have expected them to so provide. If we are to give any meaning to the small but effective phraseology actually employed, the Master's interpretation must be sustained.
Further, I do not find the concession made by Gagnebin, one of the inventors, that the patent erred in not explicitly pointing out the residual effects and form taken by magnesium when combined with impurities fatal to the Master's construction. If sufficient light may be found in the patent to guide those who subsequently take up the quest for nodular iron and omissions and errors contained therein are easily rectified or supplied by those skilled in the art the patent's disclosures are sufficient to confer on the patent its intended effect. Matheson v. Campbell, 2 Cir., 1897, 78 F. 910, 913; Balaban v. Polyfoto Corp., D.C.Del.1942, 47 F. Supp. 472, 478.
Finally, I am drawn to the same conclusion reached by the Master by virtue of the nature of the patent in suit. We are not dealing with an improvement patent but rather with a novel invention of broad pioneering dimensions. Under such circumstances, the Supreme Court has oftentimes declared that a liberality of construction should prevail. See e.g. Smith v. Snow, 1935, 294 U.S. 1, 14, 55 S. Ct. 279, 79 L. Ed. 721; Morley Sewing Machine Co. v. Lancaster, 1889, 129 U.S. 263, 273, 9 S. Ct. 299, 32 L. Ed. 715. And only recently the Second Circuit has again disclaimed any intent to strictly construe such pioneering inventions so as to deny to the inventor the fruits of his genius and labors. Georgia-Pacific Corp. v. United States Plywood Corp., 2 Cir., 1958, 258 F.2d 124. See also Samson-United Corp. v. Sears Roebuck & Co., 2 Cir., 103 F.2d 312, 315, certiorari denied 1939, 307 U.S. 638, 59 S. Ct. 1093, 83 L. Ed. 1519.
Ford argues next that if the "small but effective" claims are not to be limited to 0.04% retained magnesium on the down side they must be invalidated as indefinite on the ground that they give only a functional description at the exact point of novelty under the authority of General Electric Co. v. Wabash Appliance Corp., 1938, 304 U.S. 364, 58 S. Ct. 899, *558 82 L. Ed. 1402. But the exact point of novelty is not 0.04% of retained magnesium as contended by Ford. Instead, giving the patent the broad construction to which it is entitled the exact point of novelty must be regarded as the occurrence in cast iron, as cast, of spheroidal graphite due to the retention of magnesium. As so construed, the disclosures are set forth with sufficient definiteness and clarity to meet the test imposed by the patent laws.
The patent as the Master found is directed to the usual run of base iron encountered in foundries (M.R. p. 22) in which cases a minimal amount of 0.04% retained magnesium is suggested. It teaches further that if other irons are used adjustments must be made in the amount of added magnesium and retained magnesium depending on the quantity of impurities contained therein. This is sufficient. Where a patent contains a variable the description is sufficient if it sets forth the proportions proper under ordinary circumstances and points out the direction in which they must be varied when circumstances are changed. Consolidated Safety Valve Co. v. Crosby Steam-Gauge & Valve Co., 1885, 113 U.S. 157, 177, 5 S. Ct. 513, 28 L. Ed. 939; Burke Electric Co. v. Independent Pneumatic Tool Co., 2 Cir., 232 F. 145, 147-148, certiorari denied 1916, 241 U.S. 682, 36 S. Ct. 728, 60 L. Ed. 1234. In such cases some experimentation is unavoidable in adapting the product to the particular conditions encountered. But the patent is not invalidated on this account. It is enough if the disclosure provides sufficient direction so that persons skilled in the art can reproduce the patent without resort to independent invention. See Minerals Separation Ltd. v. Hyde, 1916, 242 U.S. 261, 271, 37 S. Ct. 82, 61 L. Ed. 286; Franc-Strohmenger & Cowan v. Arthur Siegman, Inc., 2 Cir., 1928, 27 F.2d 785. Judge Learned Hand observed in Burke Electric Co. v. Independent Pneumatic Tool Co., supra, 232 F. at page 148:
"It is as if a chemist were directed to add enough of an element to secure precipitation. Such a recipe would be an absolutely accurate guide to the result though the quantity varied with temperature or atmospheric humidity. What men need is a path to the goal; they will not be curious of the country it traverses."
And in Gillman v. Stern, 2 Cir., 1940, 114 F.2d 28, 32, certiorari denied 1941, 311 U.S. 718, 61 S. Ct. 441, 85 L. Ed. 468, we find the same recurring theme. "[A]ll that specifications need ever do is to show how to exploit the invention."
After a thorough reading of the patent, I am convinced that the area of experiment suggested is sufficiently circumscribed by the specification. Indeed, it is difficult to conceive of a patent of the kind here involved giving more than a starting point at which the optimum amount of spheroids could most uniformly be achieved. To have required the patentees to set out with particularity the critical minimum of retained magnesium necessary to induce the proper number of spheroids in each and every type of cast iron and under all conditions would seem unduly restrictive and could serve only to invite evasion by those who seek to misappropriate the substance of the invention. See Lever Bros. v. Procter & Gamble Mfg. Co., supra.
In Georgia-Pacific Corp. v. United States Plywood Corp., supra, the court after noting that the courts themselves have introduced an element of uncertainty into patent administration by invoking the doctrine of equivalents stated:
"It would be anomalous if this Court were to strike down a patent because the inevitable area of uncertainty was created by the language of the specifications and claims rather than by judicial application of equivalency doctrine" 258 F.2d at page 137.
I find the Second Circuit reasoning apposite to the case at bar and the "small but effective" claims must be taken to mean exactly what the phrase naturally imports and in the light of the circumstances *559 present here such phraseology is not to be invalidated for vagueness.
IV. Mack Defense
At the hearings before the Special Master Ford cited numerous examples of the prior art as allegedly anticipating the patent in suit. It placed its chief reliance, however, on certain valve inserts manufactured by the Mack Truck Company and its affiliates. After a searching review of Mack's activities in the cast iron field, the Master found that anticipation in law had not been established. It is to this determination that I now address myself.
Mack's history shows that for many years it had sought a satisfactory alloy for use in valve seat inserts for its truck engine blocks. These inserts, which were shrink-fitted into cylindrical recesses in the cast iron block, had to be hard and resistant to heat and wear.
In 1933 Mack was using a 40% nickel-iron alloy in the manufacture of these inserts. The castings made from this alloy, however, were often too hard and brittle for machining and proved unsatisfactory for Mack's purposes. Day, Mack's principal metallurgist during this period, upon examining these inserts, attributed their brittleness to the presence of massive carbides and to excessive porosity. He decided that impurities, principally oxides, both solid and gaseous, accounted for the presence of these defects and sought to neutralize them through the action of a "deoxidizer" or a "degasifier." For this purpose he settled on an aluminum-magnesium additive on which he executed an affidavit of invention on May 3, 1933. Before production was terminated in 1948 over half a million inserts for new truck engines alone were produced in this manner.
The Master analyzed representative samples of the Mack inserts and concluded that roughly one-half had graphite structures equal to or superior to the poorest depicted in the patent in suit. He found, however, that the spheroidal formations could not be exclusively accounted for by the retention of the magnesium in the as cast state. Other factors present in the Mack process, such as annealing, nickel content, aluminum content and small section size also contributed in part to the formation of spheroidal iron. Accordingly, though 50% of the Mack inserts possessed structures comparable to those of INCO's product, in only a considerably smaller number could such properties be attributed directly to the retention of magnesium. Ford would have me attach little significance to these figures. It argues that many of the Mack inserts, though containing smaller numbers of spheroids than suggested by the patent, did exhibit changes in graphite structure attributable to the retention of magnesium and that these changes produced, though to a lesser degree, the same desirable characteristics present in nodular iron. Ford's position is that, the difference in result between INCO and Mack being of degree only, Mack's art must, therefore, be regarded as completely and effectively foreshadowing INCO's subsequent disclosure. Accordingly, Ford excepts to the Master's conclusion that any anticipation by Mack was partial at best. Ford's approach is not wholly without merit. However, it must be noted that the rejection of the Mack defense does not turn on the percentage of overlap between the INCO and the Mack products. It rests on far firmer ground. For, as the Master properly found, whatever Mack's results, they were unappreciated and unrecognized and thus insufficient in law to invalidate an otherwise valid patent. This conclusion finds ample support in the record. Mack sought to add only sufficient magnesium to eliminate, through degasification, the porosity which it considered the source of its problem. This is evidenced by the test to which Mack subjected its inserts. An insert from each heat was placed in a V-block and struck with a heavy hammer. If it bent without breaking it was satisfactory; if it cracked, the heat was rejected. This test resulted in the acceptance of inserts containing wide variations in retained magnesium and with graphite shapes ranging from flake to *560 substantially spheroidal. Inserts containing predominantly flake structures presumably had few, if any, of the properties attributable to spheroids. They were satisfactory to Mack, nonetheless. Thus while it is true that, at least part of the time, Mack achieved an iron with improved properties due to alteration in graphite shape, these properties were not necessary for Mack's purposes and, indeed, were not the inevitable result of Mack's process. So long as its bend test indicated that brittleness had been reduced, and at least in the inserts with minimal graphite changes this could only have been due to degasification, Mack rested content.
Further, Mack either failed to recognize or was indifferent to the presence of whatever nodular iron may have resulted from its process. Although the industry had long sought spheroidal graphite, Mack never commented on the presence of spheroids in the iron with which it was working. Nor did Mack use its available equipment to test for the full range of properties attributable to the production of spheroids. Had Mack noticed any alteration in graphite shape, it is likely that it attributed this to its annealing process, and not to the retention of magnesium. Thus the production of magnesium induced nodular iron by Mack was purely a matter of chance and not the inevitable result of its process.
From the foregoing, it is clear that the occurrence of retained magnesium in the Mack inserts in quantities sufficient to form nodular iron must be regarded as accidental and unrecognized under the rule of Tilghman v. Proctor, 1880, 102 U.S. 707, 711-712, 26 L. Ed. 279. In that case, the Supreme Court, in considering whether the prior art anticipated a process patent for separating glycerin from fat acids, stated:
"if the acids [in the prior art] were accidentally and unwittingly produced whilst the operators were in pursuit of other and different results without exciting attention and without its ever being known what was done or how it had been done, it would be absurd to say that this was an anticipation of Tilghman's discovery."
It is undeniably true, as Ford asserts, that a prior use of a product deliberately created may constitute an anticipation, though the full benefits accruing therefrom may not be fully appreciated or even recognized. The prior user may have been in complete ignorance of the scientific phenomenon involved. In law, it is anticipation nevertheless. Smith v. Hall, 1937, 301 U.S. 216, 57 S. Ct. 711, 81 L. Ed. 1049; Celite Corp. v. Decilite Co., 9 Cir., 96 F.2d 242, 248 certiorari denied 1938, 305 U.S. 633, 59 S. Ct. 101, 83 L. Ed. 407; Vitamin Technologists, Inc. v. Wisconsin Alumni Research Foundation, 9 Cir., 1944, 146 F.2d 941, 948-949 certiorari denied 1945, 325 U.S. 876, 65 S. Ct. 1554, 89 L. Ed. 1994. Nor is anticipation avoided by the adoption of a new use for an unchanged product or process. Pennsylvania Railroad v. Locomotive Truck Co., 1884, 110 U.S. 490, 494, 4 S. Ct. 220, 28 L. Ed. 222; Ansonia Brass & Copper Co. v. Electrical Supply Co., 1891, 144 U.S. 11, 12 S. Ct. 601, 36 L. Ed. 327; H. K. Regar & Sons v. Scott & Williams, 2 Cir., 1933, 63 F.2d 229, 231. Thus if INCO had done no more than appreciate and exploit the full potentialities in a product already developed by Mack, or had merely found a new use for Mack's product, then Ford's position would have to be upheld.
Such, however, is not the situation in the instant case. Mack was not merely ignorant of the scientific phenomena underlying nodular iron. It is not that Mack knew it had a new iron but failed to appreciate that its unique properties were due to an alteration in graphite shape. Not only was Mack unaware that it had created a new iron but in fact Mack never consciously pursued the product here in question for any purpose. Nor did it ascribe any meaningful function to retained magnesium. Mack sought merely a means for the degasification of iron. Thus General Electric Co. v. Jewel Incandescent Lamp Co., 1945, 326 U.S. 242, 66 S. Ct. 81, 83, 90 L. Ed. 43 on *561 which Ford relies so heavily, is inapposite here. There, the product of both the prior art and the patentee was a double-etched glass surface. The patentee employed the etching process in order to add strength, whereas the objective of the prior art was to diffuse light. In finding anticipation the Court wrote:
"If A without mentioning the element of strength patented a bulb which was extra strong, B could not obtain a patent on the bulb because of its strength, though he was the first to recognize that feature of it."
In General Electric, the product, double-etched glass, was deliberately sought by both the prior art and the patentee. Thus the patentee did no more than discover a new use for an unchanged product. This is to be contrasted with the facts of the instant case. Here Mack never sought to manufacture nodular iron, the product which INCO has discovered. Indeed it was unaware that it had done so and had no idea of the significance of retained magnesium. Where the allegedly anticipating product was produced merely by chance and never recognized nor appreciated, one who later discovers and recognizes the product may patent it. Even where the allegedly anticipating product was deliberately sought and obtained by the prior art the discovery of a new use need not be invalidated for anticipation. As Judge Learned Hand observed in H. K. Regar & Sons v. Scott & Williams, 2 Cir., 1933, 63 F.2d 229, 231:
"When old devices are changed at all, the change may be dictated by a new conception, which it took originality to conceive. Strictly, the old device is not then put to a new use; the new use begets a new device. In such cases it requires but little physical change to make an invention."
See also Eibel Process Co. v. Minnesota & Ontario Paper Co., 1923, 261 U.S. 45, 43 S. Ct. 322, 67 L. Ed. 523.
Thus in Georgia-Pacific Corp. v. United States Plywood Corp., 2 Cir., 1958, 258 F.2d 124, the Court held that a patent which teaches that edge effects and checking of plywood could be materially reduced by striating the panel with multiple and alternating grooves cut to a substantial depth, is not anticipated by a prior art which employed similar but shallower grooves to create a decorative effect in shingles. There the prior art had undoubtedly improved strength to some degree through striation, but had not appreciated this quality. The Court wrote:
"Benefits incidentally and accidentally accruing in the products of the prior art do not necessarily negate invention in a change in degree when the purpose is different and the results new and useful" 258 F.2d at page 132.
Certainly where the patented product was not even deliberately sought by the prior art it would follow a fortiori from the Georgia Pacific case that anticipation cannot be predicated on such a tenuous relationship.
The Tilghman rationale was applied to a product patent in Pittsburgh Iron & Steel Co. v. Seaman Sleeth Co., 3 Cir., 1917, 248 F. 705, 708-709. There the Court, in finding that the plaintiff's patent on an iron alloy known as "Adamite" had not been anticipated by prior alloys of similar analysis, wrote:
"If any one of the alleged anticipating alloys was Adamite, that fact, so far as the record shows, was not known to those who produced it or used it, and not being recognized as a new product with its distinctive characteristics, its production was purely an accident without profit to the art and without value as an anticipation. We are satisfied, therefore, that Adamite has not been anticipated by alloys, which, while accidentally of the same analyses were not shown to be the `article of manufacture' of the patent."
*562 It should be noted that in the Pittsburgh Iron & Steel case, as in the instant case, those who produced the allegedly anticipating alloys were aware that through their processes they had made some improvements in the iron.
The fallacy of Ford's position lies in its over-simplification and erroneous equation of the objectives sought by Mack and those sought by INCO. The pioneering technology contributed by the patentee has been sometimes characterized as ductile iron. It follows, Ford argues, that the objective of the patent was to achieve a more ductile iron. Ford would have me reject the Master's finding substantiated by the record that Mack's bend test was "designed to test porosity, not the properties which the art hoped to findand ultimately foundin nodular iron" (M.R. p. 44). Instead Ford asserts that the bend test was a test of ductility, not of porosity. According to Ford, then, the objective in adding magnesium in both the Mack art and the patent was to obtain greater ductile properties in iron. This is the crux of Ford's position. In both practices, it argues, magnesium was added to iron, the objective was the same, the results similar. Ergo Mack anticipated INCO.
However, the real point of novelty in the patent was not improved ductility, but the achievement of spheroidal graphite in the as cast iron by the retention of magnesium, thereby enabling each particular type of iron to exhibit the true properties of its matrix, whatever the matrix properties might be, with a minimum of interference from the graphite. The resulting product would be ductile only if the original matrix was ductile. It is difficult, therefore, to equate Mack's limited objective, even if we assume it to be improved ductility, with the underlying desideratum of the patent.
But even if improved ductility were the ultimate objectives of both the Mack and INCO processes, in law, there could still be no effective anticipation. Mack sought the solution to its production problems through the use of a degasifier in this case a magnesium additive. If ductility was improved, such results were attributed by Mack's metallurgists to the removal by the magnesium of impurities, and not to magnesium induced changes in the microstructure.
The underlying distinction between the two arts is obvious. Mack sought to obtain its objectives through the addition of magnesium; INCO through the addition and retention of magnesium. Mack's concern was with the magnesium that was consumed or combined; INCO's concern was with the magnesium that was retained in effective form.
The Patent Law seeks to reward those who teach the public how to perform, process or construct things which the public theretofore was unable to do because of insufficient information. See 1 Walker, Patents § 51 (Deller Ed. 1937). If a product or process is accidently produced and unrecognized by the prior art, it is unlikely that the opportunities which it presents would be within the reach of the general public. Gayler v. Wilder, 1850, 10 How. 477, 51 U.S. 477, 13 L. Ed. 504.
In the instant case, the art had sought for many years to unlock the secrets of nodular iron. The Mack process was not directed to this end. Indeed, Mack was unaware that it was at the threshold of a bold, new pioneering discovery. How else can we explain its complete failure to exploit the potential in nodular iron? *563 This is especially striking when contrasted to the wide acceptance by the industry of INCO's nodular iron. Mack could not teach what it did not know. Thus, as the Master noted, it is unlikely that one skilled in metallurgy could find in the Mack art the key to ultimate success. If, as Ford charges, the patent was a mere application and adaptation of what to one skilled in the art was obvious from the Mack process, we can be certain that the introduction of nodular iron would have been an accomplished fact long before the INCO patent.
In the last analysis, the issue of anticipation as well as invention is to be determined by how much new information the patentee has contributed to the art. Metallizing Engineer Co. v. Kenyon Bearing & A. P. Co., 2 Cir., 153 F.2d 516, certiorari denied 1946, 328 U.S. 840, 66 S. Ct. 1016, 90 L. Ed. 1615. In this case the patentees must be credited with the discovery of nodular ironan accomplishment of the highest order. The fruits of this invention should not be denied them merely because Mack with more ingenuity, insight, research and persistence might have gleaned such information from its own inserts.
Prior to INCO's public announcement concerning its invention Ford was aware of the desirability of controlling graphite shape in the production of cast iron for its crankshafts. But it was not until August 1950, after the INCO announcement, that Ford established a pilot line for casting nodular iron crankshafts. In 1952 Ford adopted a technique for producing nodular iron crankshafts known as Process I. The shafts, produced both on the pilot line and under Process I, contained, as the Master found, from 0.036% to 0.09% retained magnesium, their graphite structures being almost completely spheroidal. On August 24, 1952, ten days after the present suit was instituted, Ford's Process I production ceased in accordance with instructions issued by Ford's Legal Department. Indeed, these instructions were to the effect that no crankshafts containing more than 0.035% retained magnesium were to be used. Thereafter, Ford relied exclusively on its Process II production which it had commenced in June 1952. Under this process, nodular iron shafts were produced with magnesium residuals averaging 0.02%.
Nodular iron with low magnesium residuals was achieved by Ford through the use of base iron from which substantial amounts of sulfur and oxygen had been previously removed. The Master found the removal of these impurities to be the equivalent of the magnesium that would otherwise have been required and that shafts manufactured under it infringed claims 1, 4, 13, 15 and 17 which specify magnesium minimums of about 0.04%.
Ford excepts to this finding and argues that the patent, in terms, confines itself to minimum magnesium residuals of 0.04%, leaving in the public domain metals with magnesium residuals below that. Indeed, it is asserted that the patent disclaims irons with magnesium residuals below 0.04%.
The doctrine of equivalents upon which the Master based his conclusion is old in the law. Winans v. Denmead, 1854, 15 How. 330, 56 U.S. 330, 14 L. Ed. 717. Under it, infringement cannot be avoided by the substitution of one element in a patent for another, where, as here, the two perform substantially the same function in substantially the same way. Graver Tank & Mfg. Co. v. Linde Air Prod. Co., 1950, 339 U.S. 605, 70 S. Ct. 854, 94 L. Ed. 1097.
Ford's arguments overlook the two distinct functions performed by the magnesium retained in the iron. First, as I have already stated, a certain amount combines with impurities, thus counteracting or neutralizing them. The more impurities present, the more magnesium *564 will be required. This combined magnesium is unavailable for the second function of inducing the formation of spheroids. As a result, the amount of retained magnesium necessary to produce a given percentage of spheroids varies with the quantity of impurities present. Therefore, the quantum of retained magnesium which is present in the INCO and absent in the Ford iron did not produce spheroids. It merely removed impurities in order to permit other magnesium to act upon the graphite structure, a function rendered unnecessary by Ford's use of purer iron.
The Master correctly found that the patent, describing as it does a new product long sought by the industry, is entitled to the broad range of equivalents usually afforded pioneering patents. Continental Paper Bag Co. v. Eastern Paper Bag Co., 1908, 210 U.S. 405, 28 S. Ct. 748, 52 L. Ed. 1122; Morley Sewing Mach. Co. v. Lancaster, 1889, 129 U.S. 263, 9 S. Ct. 299, 32 L. Ed. 715; Treibacher-Chemische Werke etc., v. Roessler & H. C. Co., 2 Cir., 1914, 219 F. 210. In any event, no broad reach is needed to equate the use of magnesium to remove impurities with the use of iron from which the impurities have already been removed. As we have seen, the purity of the iron and the amount of magnesium vary inversely. Musher Foundation v. Alba Trading Co., 2 Cir., 150 F.2d 885, 887, certiorari denied 1945, 326 U.S. 770, 66 S. Ct. 175, 90 L. Ed. 465, though involving a process and not a product patent, provides a close parallel. There the Court, per Judge Learned Hand, wrote:
"The undisputed testimony was that temperature and time are correlative factors * * * i. e. the time may be shortened, if the temperature is raised, and must be lengthened, if it is not. Claim One selects, as the combination, `a slightly elevated temperature' and `a short time'; the defendant departs from this by using no heat but allowing the infusion to go on for three days at least. When dealing with such a claim, it would certainly be wrong to divorce heat and time into two separate elements; and to hold that, because the defendant left out heat, but greatly lengthened time, it omitted one of the constituents of the claim."
The Court in Musher found infringement under the doctrine of equivalents.
As noted earlier in my discussion of the "small but effective" claims, the Master found that one skilled in the art would read the percentages in the patent as applying only to the usual run of base irons and would realize that "by purifying his iron, he could successfully use less than 0.04% retained magnesium." Thus the equivalent used by Ford, having been directly taught by the patent, was not discovered by Ford as a result of independent research. Indeed, the Master found that the techniques used by Ford in its Process II were widely known in the art before the patent in suit issued.
However, even were I to reject these findings, I would still hold that the doctrine of equivalents applies. In the case of a pioneering patent such as the one in suit, prior knowledge is not required. Morley Sewing Mach. Co. v. Lancaster, 1889, 129 U.S. 263, 9 S. Ct. 299, 32 L. Ed. 715; see also Priebe & Sons Co. v. Hunt, 8 Cir., 188 F.2d 880, appeal dismissed 1951, 342 U.S. 801, 72 S. Ct. 92, 96 L. Ed. 607.
The Master's finding that the percentages given in the patent would be read with the qualification that they apply only to normal base irons serves to negate Ford's argument based on dictum *565 in the old case of Treibacher-Chemische Werke v. Roessler & H. C. Co., 2 Cir., 1914, 219 F. 210, 212, that by stating that 0.04% was essential the patent disclaimed the ranges below that. If there were disclaimers, they were applicable only to melts using normal base iron.
Finally, my finding of equivalence is in no way inconsistent with my rejection of the Mack defense. Ford has urged that if preliminary deoxidation by wholly unrelated means is the equivalent of magnesium retention for purposes of infringement, then preliminary deoxidation by magnesium itself is the equivalent of magnesium retention for purposes of anticipation. Here again Ford has overlooked the dual function of the retained magnesium. Mack knowingly used magnesium only to remove impurities. There is no equivalence between that magnesium and the magnesium used by INCO to alter the graphite structure.
I come now to Ford's contention that, even if infringed, the patent is unenforceable because INCO has misused it "by a licensing system which discourages application for patents and which virtually forecloses the development of any competitive system of licensing." The misuse defense was rejected by the Master. With this conclusion I am in agreement.
Under INCO's license agreements, licensees grant licenses to INCO on any discovery or invention of the licensee involving "an improvement on any product or process covered by a patent included in `INCO's Patent Rights,' or any new or improved process, `Addition Agents' or `Devices or Apparatus' for the production of any such product or improvement thereon * * *" Under the terms of the agreement, INCO sub-licenses the improvement to all licensees of its own patents at no additional royalty. Upon termination of the agreement, due to the expiration of INCO's own patents, all licensees may continue to practice the improvement patents royalty free. In 1955, the form of the agreement was altered to make the license-backs non-exclusive, i. e. the licensees retain the right to sub-license their own improvement patents to others.
That license-back provisions are not illegal per se was established beyond cavil by Transparent-Wrap Machine Corp. v. Stokes & Smith Co., 1947, 329 U.S. 637, 67 S. Ct. 610, 91 L. Ed. 563. True it is, that license agreements will not be upheld where they are used in violation of the anti-trust laws to perpetuate control over an industry long after the expiration of the basic patents or where "through * * * patent pools or multiple licensing agreements the fruits of invention of an entire industry might be systematically funneled into the hands of the original patentee." Similarly such agreements which tend to stifle research are antagonistic to the underlying policies of the patent laws.
Thus, license-backs and pooling agreements are illegal when used to fix prices. United States v. General Electric Co., D.C.S.D.N.Y.1948, 80 F. Supp. 989; United States v. General Electric Co., D.C.N.J.1949, 82 F. Supp. 753; United States v. Line Material Co., 1948, 333 U.S. 287, 68 S. Ct. 550, 92 L. Ed. 701, or to allocate markets and fields of manufacture. Hartford Empire Co. v. United States, 1945, 323 U.S. 386, 65 S. Ct. 373, 89 L. Ed. 322; United States v. Imperial Chemical Industries, D.C.S.D.N.Y.1952, 105 F. Supp. 215. But there is no evidence here of such restraining uses of the INCO license-backs and in their absence the license-backs are a reasonable utilization of INCO's patents. See Cutter Laboratories v. Lyophile-Cryochem Corp., 9 Cir., 1949, 179 F.2d 80; United States v. Birdsboro Steel Foundry & Mach. Co., D.C.W.D.Pa.1956, 139 F. Supp. 244; United States v. E. I. duPont deNemours & Co., D.C.Del.1953, 118 F. Supp. 41, affirmed 1956, 351 U.S. 377, 76 S. Ct. 994, 100 L. Ed. 1264.
A fair reading of the agreements discloses that the licensees are required to license back only patents which could not be exploited without risking infringement of INCO's basic patent. I fail to *566 see how INCO's control of the industry is thus substantially increased. Stokes & Smith Co. v. Transparent-Wrap Mach. Co., 2 Cir., 1947, 161 F.2d 565; United States v. E. I. duPont deNemours & Co., supra. Anyone wishing to practice one of these improvements without the risk of infringement would have to come to INCO for a license regardless of who held the rights to the improvement. In effect then, much of the control over the improvements during the life of the basic patent which INCO gains by the license-backs it already has by virtue of its basic patent. On the other hand the INCO agreements assure that improvements will be disseminated throughout the long list of INCO's licensees without extra charge. The net effect may well be to increase rather than to decrease competition. See Transparent-Wrap Machine Corp. v. Stokes & Smith Co., supra.
Upon the expiration of INCO's own patents its licensees will be permitted under the terms of the agreements to practice all of the acquired improvement patents royalty free. In this respect the INCO agreements are more liberal than that in the Transparent-Wrap Mach. Corp. case, and are unlikely to result in any unreasonable restraint of competition.
I find Ford's allegations further negated by the fact that INCO's licenses are offered to all qualified applicants at reasonable rates and on a non-discriminatory basis. Indeed, it was to INCO's vital interest to secure the widest dispersion of its patented knowledge. For, as the Master emphasized, INCO is not a producer of nodular iron. In this field it lives by its royalties.
Ford, however, argues that INCO's status as a non-producer is irrelevant. It reasons that INCO is in the business of licensing "know how" and that it is this business that INCO has monopolized. I find this argument unconvincing. Certainly INCO has monopolized the business of licensing the practice of its patents. But that is a legal monopoly within the four corners of the patent. Further, the license-backs do not give INCO's nodular iron the extra leverage in its competition with rival metals and processes that they would first appear to afford. I realize that every improvement added to the INCO package makes that package more attractive vis-à-vis other licensors' wares. But, to a great extent, this is inevitable no matter who actually possesses the improvement patent.
I need devote little discussion to the allegation that the license-backs have stifled research. The Master found that considerable research has been conducted by licensees and that, any claim that absent the agreement, there would have been greater research is mere conjecture. This conclusion is supported by the record.
VII. Notice of Infringement
There remains one more question for my determination, which requires comment: the date from which INCO may recover damages. The controlling statute, 35 U.S.C. § 287, provides that where a patentee's product is not marked in the manner prescribed he may recover damages only
"on proof that the infringer was notified of the infringement and continued to infringe thereafter, in which event damages may be recovered only for infringement occurring after such notice."
It being conceded that the requisite markings were absent, the Master found that actual notice of infringement was given Ford on March 11, 1952 and that damages are to be computed from that date. Both sides have excepted to this finding.
Ford contends that there is no evidence that it received the statutory notice prior to the institution of suit, and that, accordingly, damages did not begin to accrue until that date. INCO, on the other hand, asserting that Ford knew it was infringing, would have damages computed from the day Ford began commercial production of nodular iron crankshafts. I hold that the Master's finding on this point, having sufficient evidentiary support, is not "clearly erroneous" and must be sustained.
*567 The facts, as found by the Master, are that on March 11, 1952, during negotiations between the parties, McRae, a Ford house patent counsel acknowledged that Ford was "piling up liabilities" to INCO by its indecision whether to become a licensee under the patent in suit. Coming from a trained patent counsel, who must have known that notice is necessary for the accruing of damages for infringement, such a statement was an admission of the receipt of such notice. "When one acknowledges for his adversary that the adversary is claiming infringement, the law most certainly does not compel the patent owner to repeat it more explicitly." Livesay Window Co. v. Livesay Industries, 5 Cir., 1958, 251 F.2d 469, 475. I might add that I consider it significant that although McRae was still in Ford's employ it failed to call him to testify before the Master.
March 11, 1952, was then the latest possible time to begin the computation. It is also the earliest, since INCO has not shown that it gave the notice required by 35 U.S.C. § 287 before that date. Evidence of Ford's state of mind, i. e. that Ford knew of the patent and had the soundest possible reason to believe that INCO would regard quantity production of nodular iron as an infringement, is irrelevant. To hold otherwise would be to read the statute to say: "* * * damages may be recovered for infringement occurring after the infringer is, or should have been, aware of such infringement." This is not the law. The statutory requirement of "notice" is unambiguous. There can be no recovery for the period before the defendant is expressly notified by the patentee that it is infringing a particular patent. Smith v. Dental Products Co., 7 Cir., 140 F.2d 140, certiorari denied, 1944, 322 U.S. 743, 64 S. Ct. 1146, 88 L. Ed. 1576; Parker Rust Proof Co. v. Ford Motor Co., D.C.E.D.Mich.1928, 23 F.2d 502; Franklin Brass Foundry Co. v. Shapiro & Aronson, Inc., 3 Cir., 1921, 278 F. 435; Westinghouse Electric & Mfg. Co. v. Condit Electrical Mfg. Co., C.C.S.D. N.Y.1908, 159 F. 154; Kilgore Mfg. Co. v. Triumph Explosives, Inc., D.C.Md. 1941, 37 F. Supp. 766, 776, modified 4 Cir., 128 F.2d 444, certiorari denied Sub Nom. Faber v. Triumph Explosives, 1942, 317 U.S. 660, 63 S. Ct. 59, 87 L. Ed. 530. I need not decide how, if at all, the situation would be altered, if INCO had shown that Ford had fraudulently concealed its activities from INCO. The Master who heard the evidence and observed the witnesses found that fraud had not been established. The record substantiates this finding which is not "clearly erroneous" and must be sustained.
The status of the defendant Caswell is somewhat different from Ford's in this regard. There is no contention that it received notice of infringement prior to August 14, 1952, the date on which this suit was filed. Thus its liability cannot begin to accrue before that date.
Certain other contentions raised by the parties are completely without merit and do not warrant the full discussion afforded the foregoing points. Suffice it to say that as to those contentions, I see no commission of error and the record substantiates the Master's findings of fact and conclusions of law. It is, therefore, the judgment of this Court that:
(1) the Master's findings of fact and conclusions of law are confirmed;
(2) United States Patent No. 2,485,760 is valid as to the claims in suit;
(3) the nodular iron crankshafts manufactured on the pilot line and under Process I by defendant Ford and such shafts sold by defendant Caswell infringe Claims 1, 3, 4, 6, 9, 10, 11, 12, 13, 14, 15 and 17;
(4) those containing more than 0.045% retained magnesium also infringe Claims 2, 5 and 7;
(5) the nodular iron crankshafts manufactured under Processes II and III by defendant Ford and such shafts sold by defendant Caswell infringe Claims 1, 3, 4, 6, 9, 10, 11, 12, 13, 14, 15 and 17;
(6) defendants are enjoined from further acts of infringement;
*568 (7) an accounting for damages is directed and shall be computed from March 11, 1952 in the case of Ford and August 14, 1952 in the case of Caswell;
(8) said accounting will proceed before the Honorable Simon H. Rifkind, the Special Master previously appointed herein upon the request and consent of the parties, and the issues, if any, as to punitive or increased damages will be heard at said accounting and reported on by said Master;
(9) defendants' counterclaim for a declaratory judgment of invalidity and non-infringement is dismissed on the merits.
 The principal defendant is concededly Ford and the Master's Report made little mention of the defendant Caswell. Concluding that Caswell's status as to infringement was basically the same as Ford's, except as to the date from which damages would be computed, the Master directed himself primarily to the arguments presented by Ford. I shall do likewise.
 In addition, the Master conducted his own examination of the photomicrographs offered in evidence, as did this Court.
 See e. g. Master's comment on Schneidewind's testimony (M.R. p. 24).
 The Master divided the 15 allegedly infringed claims into the following 3 categories on the basis of their reference to retained magnesium:
1. Claims 1, 4, 13, 15 and 17, in which "about 0.04%" is stated as the minimum quantity of retained magnesium. In all but Claim 15, the maximum is stated as being 0.3%. Claim 15 sets the maximum at 0.5%.
2. Claims 2, 5 and 7, in which 0.05% is stated as the minimum quantity of retained magnesium and 0.2% set as the maximum.
3. Claims 3, 6, 9, 10, 11, 12 and 14, in which the minimum quantity of retained magnesium is defined only as a "small but effective amount" to control or effect or promote the occurrence of graphite spheroids. Maxima in these claims vary from 0.2% to 0.5%.
 This finding is well supported by the record. For example, the discovery of nodular iron enabled one licensee to produce large numbers of bearing support rings for jet engines. The licensee found that these components, which were built to withstand temperatures ranging from 700 degrees to zero, could not be successfully fabricated from any other ferrous casting material. Between 5,000 and 10,000 of these support rings were produced. This is only one of the many uses for nodular iron described in the record. There is also impressive evidence of the new metal's superior properties such as ductility, tensile strength and high yield strength, etc.
 This figure is made more significant by the fact that INCO neither made nor sold nodular iron. Thus this considerable sum was in addition to any expenditures made by licensees for the purpose of promoting their products.
 The specification refers at different points to "a critical minimum of about 0.04% retained magnesium, preferably at least 0.05% retained magnesium" (Col. 3, lines 48-50), to "the presence of about 0.04% or more retained magnesium" as one of the "essential features" of the invention (Col. 4, lines 10-16), and to "the retention of at least a critical minimum amount of magnesium in the final product" (Col. 5, lines 6-7). There is also the statement made with reference to an illustrative series of castings of a particular base composition, that 0.02% is "too little" (Col. 16, lines 68-9).
 The Master pointed to that language in the patent which referred to "borderline amounts of retained magnesium, e. g. 0.035% to 0.043%, just sufficient to initiate the occurrence of the aforementioned spheroidal or spherical form of carbon" (Col. 6, lines 42-46); the indication that measurement of the quantity of retained magnesium is difficult and imprecise, being likely to vary within about ±0.005% (Col. 7, lines 45-57); and the reference to the fact that the minimum quantity of retained magnesium will increase slightly with the carbon and/or silicon content and with the section size of the casting (Col. 6, lines 23-28).
 This finding is significant inasmuch as the allegedly infringing products were made of a cast iron containing a maximum retained magnesium content of 0.02%. Further discussion on this point is to be found elsewhere in this Opinion.
 Ford's position is that it was Ford, and not the patentees, that taught that retained magnesium was made up of undisclosed quantities of ineffective as well as effective magnesium. I think that the timing of Ford's development of nodular iron as discussed later amply refutes this contention.
 Such was the import of the testimony of Gagnebin, one of the inventors. The Master found further corroboration of this interpretaton in the testimony of Schneidewind, one of Ford's expert witnesses at the hearings.
 See also Ludlum Steel Co. v. Terry, D.C.N.D.N.Y.1928, 37 F.2d 153; Rem Cru Titanium v. Watson, D.C.1956, 147 F. Supp. 915; Pittsburgh Reduction Co. v. Cowles Electric S. & A. Co., D.C. N.D.Ohio 1893, 55 F. 301; cf. Wickelman v. A. B. Dick Co., 2 Cir., 1898, 88 F. 264; Standard Envelope Sealer Mfg. Co. v. Graywood Mfg. Co., D.C. Mass. 1922, 1 F.2d 667.
 The specification of the patent in suit makes it abundantly clear that it is the retention and not the mere addition of magnesium which is important.
"It is not sufficient merely to add magnesium to the molten bath. The presence of retained magnesium in the aforementioned product is essential in order to obtain the aforementioned structure therein" (Col. 5, lines 8-12).
 In July 1954 Process II was superseded by Process III. For our purposes, however, there are no significant differences between the two.
 The Master concluded:
"I have held, after all, that INCO'S invention was of a pioneering nature specifically because no one theretofore had discovered how to achieve spheroidal graphite. Thus, no equivalents for retained magnesium for this purpose, were known at the time this patent issued. My duty to accord this pioneering patent a broad range of equivalents is a nullity unless possible equivalents discovered after issuance may be considered" (M.R. pp. 69-70).
I find this reasoning most persuasive.