359 F.2d 739
CORN PRODUCTS COMPANY, Plaintiff-Appellant,
v.
STANDARD BRANDS, INC., Defendant-Appellee.
No. 15131.
United States Court of Appeals Seventh Circuit.
April 11, 1966.
W. Houston Kenyon, Jr., New York City, James P. Hume, Granger Cook, Jr., Chicago, Ill., for plaintiff-appellant, Edward W. Greason, William F. Noval, Robert T. Tobin, Kenyon & Kenyon, New York City, Hume, Groen, Clement & Hume, Chicago, Ill., of counsel.
John C. Butler, Chicago, Ill., Hubert Hickam, Indianapolis, Ind., Maxwell Barus, W. Philip Churchill, Albert E. Fey, New York City, for defendant-appellee, Fish, Richardson & Neave, New York City, Kirkland, Ellis, Hodson, Chaffetz & Masters, Chicago, Ill., Barnes, Hickam, Pantzer & Boyd, Indianapolis, Ind., of counsel.
Before DUFFY, KNOCH and MAJOR, Circuit Judges.
DUFFY, Circuit Judge.
The complaint herein alleged the infringement of certain claims of the Melnick and Luckmann Patent, No. 2,955,039 issued on October 4, 1960, and owned by the plaintiff. The patent in suit contains sixteen claims of which claims 1, 2 and 9 to 14 are directed to a margarine product, and claims 3 to 8 and 15 and 16 are directed to a process of preparing an edible vegetable oil blend to be used in making margarine. The alleged novelty of the margarine oil is the blending thereof which will be hereinafter described.
Plaintiff alleged that defendant infringed claims 1, 2, 11 and 14 of the patent in suit. Defendant denied infringement, alleged the invalidity of the patent claims, and by a counterclaim, sought a declaratory judgment of non-infringement and invalidity as to all sixteen claims of the patent.
Defendant's products, charged with infringing claims 1, 2, 11 and 14 of the patent, are two brands of margarine sold by defendant under the names "Fleischmann's Regular" (salted) and "Fleischmann's Unsalted."
Since oil suppliers furnished defendant with its margarine oils and this oil was already blended and ready to be made into margarine, no charge was made in the complaint that defendant had infringed process claims 3 to 8, 15 and 16 of the patent in suit.
After a trial, the case was submitted to the District Court without oral argument on March 28, 1963, upon briefs and proposed findings and conclusions filed by each party. A decision was handed down on December 31, 1964, and the findings of fact and conclusions of law were filed on the same day. No opinion was filed. To a large degree the District Judge adopted defendant's proposed findings and conclusions.
The District Court held all sixteen claims of the patent invalid. However, Conclusion No. 8 states that claims 1, 2, 11 and 14 "read on defendant's two Fleischmann margarines" but that "defendant has not infringed such claims for the reason that they are each invalid."
Plaintiff asserts that the patent in suit describes and claims a "new grocery-store margarine" and two processes of making it which, while maintaining unimpaired the butter-like physical characteristics which enable ordinary margarine to be packed in quarter-pound prints, sold in grocery stores and used in the home as butter is used, possesses the novel chemical attribute that its oil phase contains from two to three or more times as much linoleic fatty acid as is found in ordinary margarine, but without substantial increase in the saturated fatty acid thereof, so that the new margarine, in contradistinction to ordinary margarine, contains a substantial excess of natural linoleic acid over saturated fatty acids. The patent defines this excess as a "ratio in excess of 1.0" which means that the quotient arrived at by dividing the percent of linoleic by the percent of saturates is a number greater than one. This ratio is referred to as the "L/S ratio."
Margarine, to be readily saleable, must closely approximate the physical characteristics of butter as to (a) firmness at refrigerator temperatures (50° F.) which permits manufacture and wrapping in the form of quarter-pound prints (sometimes called "printability"); (b) retention of form at room temperature (70° F.) (sometimes called "stand-up" or "nonslumping"); (c) resistance to oil-staining of cartons in warm weather (80° F.), and (d) quick melting just below human body temperature (92° F.) which enables the emulsion to break up on the tongue and thus release flavors, — otherwise the product would have a greasy taste such as a kitchen shortening if spread on the tongue.
A test which is industry-accepted and used in the specification and claims of the patent in suit is called the "dilatometric method." By this method, the "Solids Content Index" ("S.C.I. values" or "SCIs") of a particular oil is determined at each of the temperatures 50°, 70°, 80° and 92° F. These four SCI values are plotted and a curve drawn. This curve, taken with the final melting point of the margarine, constitutes a concise definition of the physical characteristics and it is used in the claims of the patent in suit. A very steep slope in this curve is characteristic of butter.
Vegetable seed oils such as corn, cottonseed and soybean are limpid waterlike fluids at room and refrigerator temperatures. These oils can be hardened by a method termed hydrogenation, and thus hardened, can be used in making margarine. In this hydrogenation process, hydrogen gas is bubbled through hot oil under pressure. The oil contains a catalyst such as nickel or platinum.
Before hydrogenation, vegetable seed oils contain three principal natural components.
1) Poly-unsaturates are fatty acid radicals, which, at two or more places, along the carbon-atom chain, have a deficiency in hydrogen and are capable of adding it there. The most abundant poly-unsaturate is natural linoleic acid having two hydrogen-deficiency points, which comprises nearly 50% or more of corn, cottonseed and soybean oils. Animal fats and coconut oil, in contrast, contain only a very small percentage of linoleic acid.
2) Mono-unsaturates are fatty acid radicals that have a hydrogen-deficiency at one place. The only mono-unsaturate occurring in nature is oleic acid.
3) Saturates are fatty acid radicals which have no hydrogen deficiency point and cannot add hydrogen. They comprise about 10 to 15% of corn and soybean oils and about 23 to 25% of cottonseed oil. They comprise nearly 50% of the animal fats and almost 90% of coconut oil.
Hydrogenation has two effects in an oil. 1) Hydrogen is absorbed at the hydrogen-deficiency points of the poly- and mono-unsaturates, converting linoleic into oleic, and oleic into saturated fatty acid. 2) Disturbances are created in the carbon-atom chains such that existing hydrogen-deficiency points change in character or shift position along the chains without absorbing hydrogen, thereby creating man-made substances called "isomers" that differ in melting point and other characteristics from the natural components of the oil.
In hardening an oil for margarine use, the effort is to obtain a melting point just below human body temperature and an SCI slope about as steep as that of butter. The former is obtained by stopping when the oil's melting point reaches 95° F. The latter is achieved by what is called "selective hydrogenation", a technique which speeds the conversion of linoleic into oleic, and slows the conversion of oleic into saturates.
Plaintiff claims the invention of the patent in suit overcame the dilemma by blending together two moieties of oil of radically different character. One is a liquid unhydrogenated vegetable seed oil such as corn, cottonseed or soybean, which, at temperatures of 50° F. and above is a water-like fluid. The other, the "hard fraction", is a vegetable seed oil partially and selectively hydrogenated to a melting point higher than that required in ordinary margarine, so that no more than about 5% (but preferably none) of the linoleic acid remains, so that saturated fatty acids have not increased more than 16%, and so that the SCI curve of the hydrogenated oil has a slope (between 50° and 92° F.) of from 30 to 45 S.C.I. units.
On August 4, 1962, the Council on Foods and Nutrition of the American Medical Association, published in the American Medical Association Journal, a comprehensive report titled "The Regulation of Dietary Fat." The article advised physicians that the preferred treatment for hypercholesteremia (abnormally high cholesterol content in the blood plasma, a condition claimed to be associated with high incidence of heart attack and strokes) is a diet in which "the ratio of poly-unsaturated fatty acids to saturated fatty acids ranges from 1.1:1 to 1.5:1 (an L/S ratio of 1.1 to 1.5)". The article also recommended specific diets for use in such cases in which "Special" margarine is a mandatory substitute for ordinary margarine and butter.
Plaintiff claims the discovery upon which the invention described in the patent rests is that the particular hard fraction defined in the patent will, when combined with liquid unhydrogenated vegetable seed oil, interact therewith to form itself into a matrix capable of holding the large amount of liquid oil required to provide an L/S ratio materially in excess of 1.0%, with the combination of matrix plus-liquid possessing butter-like physical characteristics at every relevant temperature including the capacity to melt completely at human body temperature, releasing both moieties of the combination as liquids, even though the melting point of the hard fraction before it was combined had originally been higher.
Natural linoleic and other higher poly-unsaturates are referred to in the patent in suit as "essential fatty acids" because they are required for control of plasma cholesterol levels in the human body. The patent states the object of the invention is to provide a margarine "of a high essential fatty acid content" in which there will be "a high ratio of essential fatty acid to saturated fatty acid content," with "a fatty acid pattern approximating that of the liquid unhydrogenated domestic vegetable oils." The patent also states that in the new margarine "the ratio of linoleic to saturated fatty acids * * * [is] * * * in excess of 1.0" and preferably "in excess of 1.2". In Table IV and in Table VI, eighteen examples are given having L/S ratios from 1.1 to 2.8.
The patent discloses two methods of making this product. The first is known as the 1-component method (claims 3 to 8 inclusive). It consists of blending 30% to 70% of an unhydrogenated vegetable oil containing upwards of 25% linoleic acid with a vegetable oil that has been hydrogenated "selectively"; which has a melting point from about 98° to 112°, preferably 100° to 110°; its linoleic acid content has been reduced to not more than 5% and preferably to complete extinction; its saturates have not been increased more than 16% and preferably not more than 14%; and its SCI slope is between about 30 to 45 S.C.I. units.
The second or 3-component method (claims 15 and 16) is used by plaintiff in one of its margarines, but the record does not disclose that it was used by defendant.
The margarine produced by either method qualifies as "Special" margarine referred to by the AMA Council, and differs from the prior art ordinary margarines in having an L/S ratio in excess of 1.0 and also differs from the product known as "Emdee" (a canned drugstore product) in having butter-like physical characteristics. Plaintiff says this enables persons who cannot afford drugstore products for daily use to purchase in grocery stores a margarine with butter-like characteristics having an L/S ratio above 1.0 at margarine prices.
Claims 1 and 2 of the patent cover the described margarine product. Plaintiff says they are "a new composition of matter, defined in terms of the several physical and chemical properties which collectively distinguish its novelty."
At the time of the trial in October 1962, plaintiff's combined sales of "Mazola" and "Nucoa" margarines, each a "Special" margarine made according to the patent in suit, were in excess of 80 million pounds per year.
Defendant began to manufacture and sell its new "Fleischmann's Unsalted" margarine in September of 1960. The District Court found it to be a 2-component blend within the claims of the patent in suit. On September 1, 1962, defendant converted its prior art salted "Fleischmann's Regular" to the same 2-component blend then being used in "Fleischmann's Unsalted."
In 1956 and 1957, plaintiff had attempted to produce a "grocery store" margarine with high linoleic acid content by a method called interesterification. An application for a patent was filed, but it was soon apparent that interesterification failed to solve the problem because the SCI slope could not be made steep enough to match that of butter. The product was unprintable until coconut oil was added; but upon adding this oil, the L/S ratio dropped to 0.87 and the melting point rose to 103° F. giving the product an unpleasant greasy taste. Plaintiff never put its interesterified margarine on the market.
Frequent reference is made throughout this case to a prior art commercial margarine known as "Emdee" which blended 80% liquid unhydrogenated corn oil with 20% hydrogenated coconut oil. However, as the District Court found, Emdee margarine oil differed from the margarine oil claimed in the patent in suit in that it was softer at refrigerator and room temperatures to the extent it could not be printed as a commercial margarine and it was and is sold in a can as a drugstore product.
One of the principal obstacles which plaintiff necessarily had to overcome to be successful in this suit was the undenied fact that from 1937 to 1942, the John F. Jelke Company of Chicago, Illinois, manufactured and sold Good Luck margarine. The trial court concluded that claims 1 to 14, inclusive, of the patent were invalid for lack of novelty over the prior manufacture and sale of Good Luck.
During 1937 to 1942, sales of Jelke's Good Luck margarine averaged in excess of one million pounds per month. The margarine was sold in one pound prints wrapped in parchment and waxed paper. It had a Wiley melting point of 93° F. to 98° F., good stand-up quality at room temperature and melted quickly in the mouth.
Jelke purchased its component oils from vegetable oil suppliers, principally Swift & Company, Procter & Gamble (P&G) and Southern Cotton Oil Company. Jelke blended its own oil and used the blend in the making of its margarine. The exact proportions of liquid unhydrogenated cottonseed oil used in the blend were varied from time to time according to the decisions of a "Taste Committee" composed of Jelke officials.
Plaintiff does not dispute the fact that Jelke manufactured and sold over fifteen million pounds of Good Luck margarine, deliberately made with an oil phase containing about 51.2% of liquid unhydrogenated cottonseed oil.
Many of Jelke's original manufacturing records were received in evidence. One of the most important documentary exhibits was the original formula book. This book showed the amounts of each kind of oil used to make up a churn of margarine containing 2,150 pounds of oil. From 1937 through 1942, only two kinds of oil were used. One was an hydrogenated cottonseed oil identified by former Jelke employees as having been hydrogenated to a 102° Wiley melting point plus or minus 2°, by Jelke's oil suppliers.
The second oil component used during this period was usually referred to as "#22 cotton oil." This oil was identified as a refined, deodorized cottonseed oil which had not been winterized or hydrogenated and had not had anything added to it.
Plaintiff argues the evidence is not clear and convincing that Good Luck margarine had an L/S ratio in excess of 1.0 for any considerable length of time because some tank cars of liquid cottonseed oil received by Jelke had a high cloud point indicating contamination. Such contamination was probably cottonseed stearine which contained less linoleic and more saturates than normal cottonseed oil.
We do not think the argument based upon stearine contamination is convincing. Records in evidence disclosed that from November 1938 to March 4, 1940, 182 shipments of liquid cottonseed oil had been received by Jelke. In only ten were cloud points of more than 5° C. recorded. By another computation, only ten tank cars of cottonseed oil out of a total of 321 showed any appreciable contamination.
The patent in suit defines the margarine oil of the alleged invention in terms of modern analytical procedures not used when Good Luck margarine was manufactured. Hence P&G, at defendant's expense, reconstructed in its laboratories hydrogenated cottonseed oil that P&G had supplied to Jelke in the fall of 1939. This reconstructed hydrogenated cottonseed oil was then blended in the laboratory with 51% of whole liquid unhydrogenated cottonseed oil and both the hydrogenated oil component and the blend were tested and analyzed by the methods prescribed by the patent in suit.
This procedure was carried out by inter partes tests. The work was observed by representatives of both plaintiff and defendant and P&G reported the results simultaneously to both parties.
The results of these tests show that the two samples of the reconstructed hydrogenated oils and the blends made with each of these samples all fall within the arithmetical ranges defined in the most specific claims of the patent in suit. In particular, the reconstructed oils in the two blends had a ratio of linoleic to saturated fatty acids of 1.2 and 1.23, respectively.
Plaintiff argues that if Good Luck margarine had an L/S ratio in excess of 1.0, it was only for short periods of time, and says that it was "accidental and unnoticed." Of course, this argument based on the L/S ratio does not help to distinguish claims 3 to 16 which do not require a particular L/S ratio.
Plaintiff also insists that the Jelke formula was a secret process. It argues that since the formula book was customarily kept in Mr. Jelke's private safe, the formulas must have been known only to the individuals concerned, and that these formulas would have passed into oblivion if the book had not been "exhumed" in 1961 for the purpose of being offered in evidence in this case.
Plaintiff cites Minneapolis-Honeywell Regulator Company v. Midwestern Instruments, Inc., 7 Cir., 298 F.2d 36, 38, for the proposition that private knowledge and use of a secret manufacturing process, inaccessible to the public and not, in itself, a full anticipation of a patent, cannot be combined with published knowledge to establish obviousness to a man skilled in the art.
However, the record discloses that neither Jelke Company nor its successor, Lever Brothers, made any special effort to maintain secrecy with reference to the manufacture of Good Luck margarine. Many Jelke employees knew the formulas that were used from day to day, but there is no suggestion that they were sworn to secrecy or that any special precautions were taken to preserve secrecy other than keeping the formula book in a safe. Good Luck margarine was produced in an ordinary commercial factory without any special secrecy precautions.
Good Luck margarine was sold to the public in large quantities and both Swift & Company and P&G knew of Jelke's use of liquid oil. Swift used a blend of the same two cottonseed oil components in making its own "Allsweet" margarine. Furthermore, Table 3 in the Andrews and Richardson article published in 1943, gives the fatty acid analysis of Good Luck margarine.
As early as 1938, plaintiff's officials knew much about the Good Luck and Allsweet margarines on the market. They knew of the kind of hydrogenated cottonseed oil used by Swift and also knew it was blended with a substantial amount of unhydrogenated cottonseed oil. It seems apparent plaintiff could have analyzed the product for fatty acids. Andrews and Richardson did so.
For all of the above reasons, plaintiff's argument as to the secrecy of the blending formulas used in Good Luck margarine cannot be sustained.
As hereinbefore indicated, claims 3 to 16 do not say anything about the ratio of linoleic to saturated fatty acids, which is the L/S ratio so heavily stressed by plaintiff. In fact, claims 5 and 11 cover a margarine oil blend containing as little as 15% of linoleic and as much as 27% of saturated fatty acids. This would be a blend having an L/S ratio of 0.55. The most specific claims of this group, to-wit: claims 8 and 14, cover a blend having a ratio of 0.87.
The problem of making a liquid oil margarine blend containing the proper physical properties was not as difficult as plaintiff represented to the Patent Office. It is true, plaintiff did considerable experimenting, but in two weeks' time where nineteen trial blends were made, six of them were successful.
The use of liquid corn oil in the blends to obtain a high linoleic content and L/S ratio, originated with prior art Emdee margarine. It is true that claims of the patent in suit differ from Emdee, but we do not think reducing the amount of liquid oil and substituting for the hardened coconut oil a well-known hydrogenated domestic vegetable oil, amounts to an invention.
The prior art before the trial court was more pertinent than that cited in the Patent Office. This included the prior knowledge, use, manufacture and sale of Jelke's Good Luck margarine; the Black article and the Andrews and Richardson article. The Black publication taught the use of 40% liquid oil blended with the claimed kind of hydrogenated oil. Black also referred to plaintiff's expired Vahlteich patent which taught the claimed type of hydrogenation.
Plaintiff makes the usual argument as to commercial success and filling a longfelt need. In a recent case, Graham v. John Deere Co., U.S., 86 S. Ct. 684 (February 21, 1966) the Supreme Court referred to these tests saying, at page 703 "Such inquiries may lend a helping hand to the judiciary * * *. They may also serve to `guard against slipping into hindsight' * * * and to resist the temptation to read into the prior art the teachings of the invention in issue."
However, in the following paragraph of the opinion, the Court stated: "However, these factors do not, in the circumstances of this case, tip the scales of patentability." The Court then held that the patent under consideration "* * * must fall as not meeting the test of § 103, since the differences between them and the pertinent prior art would have been obvious to a person reasonably skilled in that art."
Claims 1 and 2 of plaintiff's patent are product claims which omit any reference to liquid oil, an hydrogenated oil or any kind of a blend. They call for "A margarine the total phase of which consists essentially of an edible vegetable oil having * * *" certain physical properties and chemical analysis.
Thus, claims 1 and 2 are not limited to liquid oil, to any kind of hydrogenated oil or to any kind of an oil blend. They seek to cover all margarine compositions which allegedly differ from the conventional margarines only in linoleic acid content and L/S ratio. They are broader than the claimed invention. We think the District Court was correct in holding claims 1 and 2 to be invalid and void for overclaiming the only invention described in the patent.
The critical findings of fact of the District Court cannot be said to be clearly erroneous. There is substantial credible evidence in the record to sustain them. One of these findings (No. 30), was, in substance, that over a span of two years — July 1958 to July 1960 — five groups of people working independently of each other, developed, after only a few trials, satisfactory margarine oil blends of the kind claimed in the patent in suit.
Another such finding (No. 28) is that defendant developed its own all-corn oil margarine oil blend for use in its Fleischmann's Unsalted margarine entirely independent of plaintiff and more than a year prior to the issuance of the patent in suit.
Considering the tests for patentability, i. e., novelty, utility and obviousness, under the guide lines set down by the Supreme Court in Graham v. John Deere Co., supra, and in the light of the findings of the trial court, we hold all of the claims of the patent in suit to be invalid.
The judgment of the District Court is
Affirmed.