Procter & Gamble Mfg. Co. v. Refining, Inc.

135 F.2d 900 (1943)

PROCTER & GAMBLE MFG. CO.
v.
REFINING, Inc.

No. 5027.

Circuit Court of Appeals, Fourth Circuit.

May 21, 1943.

Drury W. Cooper, of New York City (Marston Allen and Erastus S. Allen, both of Cincinnati, Ohio, and George M. Lanning, of Norfolk, Va., on the brief), for appellant.

Charles M. Thomas, of Washington, D. C. (William F. Hall, of Washington, D. C., and Barron F. Black, of Norfolk, Va., on the brief), for appellee.

Before PARKER, SOPER, and DOBIE, Circuit Judges.

SOPER, Circuit Judge.

The subject matter of this suit is United States patent No. 2,100,274 issued to Benjamin Clayton and others on November 23, 1937 upon an application filed May 2, 1931. The invention relates, as the specification states, to an improved process for refining oils, more particularly vegetable and animal oils capable of separation and refining. The invention was designed particularly for the refinement of cotton seed and similar oils, consisting principally of neutral oil, the major and more valuable component, and fatty acids together with certain impurities such as resin, gum and dye stuff. The principal objects of the invention were to reduce the time consumed in the separation of these constituents and to recover more completely the more valuable ones; and these objects were actually attained in so striking a manner that after an exhaustive trial the District Judge made the finding upon abundant evidence that Clayton had made a discovery which substantially revolutionized the art of refining vegetable oils.

The Batch or Kettle Process.

For a half century prior to the advent of Clayton, the industry employed the batch or kettle process of refining vegetable and animal oils, especially cotton seed oils. In this process crude oil and a solution of caustic soda flow by gravity from separate storage tanks into a kettle large enough to hold 60,000 pounds of the oil and 3,600 pounds of the lye, and the contents of the kettle are vigorously agitated by horizontal power driven blades so as to mix the ingredients. The quantity of caustic soda is determined after chemical analysis of the oil and is made sufficient to cause the separation of the free fatty acids and other undesirable constituents in the oils. The agitation continues until a "break" in the mixture occurs and the "foots" or soap stock begin to separate from the neutral oil. Heat is then applied by steam passed through coils in the mixture and through a steam jacket in the bottom of the kettle and the agitation is continued until a definite break between the oil and the soap stock takes place, usually at a temperature from 100 to 140° Fahrenheit. Then the agitation stops and the soap stock settles to the bottom. After the settling has reached the proper stage, the refined neutral oil is drawn off through a decanting pipe at different layers because the upper layers are freer from the soap stock; and finally the soap stock is allowed to fall by gravity into a tank below the kettle. The process is not continuous, but is applied to separate batches of oil.

Normally from 8,000 to 8,400 pounds of soap stock are collected from a batch of oil of 60,000 pounds. The soap stock, however, consists, to the extent of 30% or 2,500 pounds in weight, of entrapped neutral oil, which is the more valuable product, and in addition a substantial amount of neutral oil is lost through saponification caused by the exposure of the oil to the caustic during the long application of heat. Attempts were made in the industry to salvage a part of the entrapped neutral oil but they were abandoned in the 1920's as commercially impracticable.

The time ordinarily consumed in refining a batch of 60,000 pounds of neutral oils runs ordinarily from 12 to 24 hours, although the process has been performed in 8 or 9, according to some witnesses. The amount of oil lost and the time consumed in the process were recognized as defects characteristic of the method, and efforts were made to correct them.

Clayton Process.

Clayton declared in his application for the patent that his object was to solve these problems and that he had done so by employing a continuous process which was accomplished in a few minutes without substantial loss of the neutral oil in the foots or soap stock. Instead of a large batch of oil, as in the kettle process, small quantities of crude oil and caustic solution are mixed in a tank and the mixture is caused to pass through a pump where further mixing occurs; and then the moving stream passes through a heated coil wherein the refining and conditioning reactions are completed and the emulsion is made ready for separation in a centrifuge to which it is then conducted. The whole operation is continuous and is completed in a few minutes as compared with the "hours or days" required in the kettle process.

The streams of oil and caustic are measured and mixed in predetermined quantities and synchronized so that each increment of oil in the continuous process receives the proper increment of caustic and the uniformity of the admixture is assured before the heater is reached. The result is that the free fatty acids are neutralized with substantial completeness and the foots or soap stock is produced before the conditioning in the heater takes place. The emulsion continues to flow uninterruptedly under pressure from the mixing zone to the heating or conditioning coil, and then passes through the coil in which the temperature progressively rises, and the refining reactions are completed.

The patent states that coordination should be maintained between the constant volume of input, output and the volume of heat applied so as to establish the proper velocity and turbulence of the oil in order to ensure the diffusion of the alkaline solution and the action thereof upon every particle and globule of the oil. By these means, the experts explain, the chemical processes begun in the mixing zone are brought to completion in the coil. The emulsion is broken and the material is conditioned for the subsequent centrifugal separation because the particles are made larger by agglomeration. The viscosity of the oil is reduced and the foots are softened so that they can flow together. Claim 2 of the patent directs that the mixture be passed through a heated conduit to raise the temperature to a degree which will facilitate centrifugal separation, and the evidence shows, as the District Judge found, that sufficient information is found in the patent to guide one skilled in the art to adopt the process to the different grades and qualities of oil encountered in commercial refining, precisely as is done in the kettle process in which higher temperatures are usually employed.

The final step in the process consists in propelling the mixture under pressure at a uniform rate of feed into a centrifuge where the neutral oil is separated from the foots or soap stock precipitated by the alkali treatment in the coils.

Commercial Success.

The District Judge found from the evidence that when the Clayton process is used in the commercial refinement of vegetable oils, the total time is reduced from the hours consumed in the kettle process to minutes, and that a saving is made of 20 to 30 per cent of the neutral oil entrapped and lost in the soap stock in the kettle process. Naturally the process has met with great favor in the trade. The majority of the cotton seed oil refiners, including Swift & Company, Armour & Company, Lever Bros. Co., Simonin's Sons, Inc., Best Foods, Inc., Corn Products Refining Company and other important manufacturers have taken licenses and paid substantial royalties under the patent.

Interference Proceeding between Clayton and James.

The Procter & Gamble Manufacturing Company, the defendant-appellant in the pending case, has also paid tribute to the new process by adopting it and using it continuously since June, 1932, after an experimental or demonstration unit, installed in its plant by the Sharples Specialty Company, showed a saving of 30 per cent as compared with the batch or kettle process. At that time the defendant seems to have assumed that the invention had been made by Edward M. James and took a license from Sharples, his employer and assignee. James had in fact hit upon the process as the result of study and experiments performed in Sharples' employ, and had filed an application for a patent on October 6, 1931, six months after Clayton's filing date. As a result the Patent Office placed the two applications in interference in the summer of 1932 and there ensued a long and active contest lasting five years in which the leading attorney for the defendant in the pending case represented the James' interests. The Patent Office tribunal, and finally, the Court of Customs and Patent Appeals in a decision rendered June 21, 1937 (James v. Clayton, 90 F.2d 337, 24 C.C.P.A., Patents, 1329), awarded priority to Clayton. It was held that James was the first to conceive the invention, but Clayton was the first to reduce it to practice.

Clayton notified the defendant of the interference proceeding and offered the defendant a license under the patent in January, 1933, subject to the payment of a running royalty. The defendant was familiar with the Clayton process from an examination that took place in the plaintiff's laboratory in December, 1932, but refused the offer of a license and chose instead to purchase equipment for practicing the process from Sharples and to accept from Sharples in May, 1933, a license without payment of royalty to operate any of Sharples' existing patents or any patents that might thereafter be acquired. The Sharples equipment and process closely approximated the Clayton method. In 1938, a year after the issuance of the Clayton patent, the defendant, upon the advice of counsel, modified its equipment in a minor particular but retained the substance of the Clayton invention. The District Judge found that both of the accused processes, called herein Process A and Process B, infringed the patent in suit.

Defendant's Contention That the Scope of the Clayton Patent Was Improperly Extended by Amendment in October, 1937.

The defense is based chiefly on the contention that the patent is invalid by reason of certain prior patents and prior practices in the art, and to a less degree, upon a denial of infringement. In addition, running through the entire argument of the defendant is the assertion that the specification of the patent as originally presented to the Patent Office did not relate at all to the refinement of oils by an alkaline solution so as to neutralize the fatty acids and form soap stock, but was confined to a so-called degumming operation for the removal of gums and coloring matter which otherwise would be lost. It is said that the idea of removing the fatty acids in order to refine the oil was first introduced by Clayton by an amendment to the specification allowed on October 26, 1937, long after the defendant had publicly adopted the process on which the charge of infringement is based. Therefore, it is contended that the claims of the patent are invalid (1) because they were not supported by the original specification and (2) because the defendant had made public use of the patented process as disclosed by the amendment more than two years before it was filed. To support this argument reference is made to Schriber-Schroth Co. v. Cleveland Trust Co., 305 U.S. 47, 573, 59 S. Ct. 8, 83 L. Ed. 34; Mackay Radio & Telegraph Co. v. Radio Corp., 306 U.S. 86, 618, 59 S. Ct. 427, 83 L. Ed. 506; and Muncie Gear Works v. Outboard Marine & Mfg. Co., 315 U.S. 759, 62 S. Ct. 865, 86 L. Ed. 1171.

These contentions would be serious if supported by the evidence, but an examination of the original specification and of the relevant dates discloses that the charge is entirely lacking in substantial foundation. The original specification plainly states that "the principal object of the invention is to reduce the time period in the separation of the various constituents of oils, such as the `foots' for cotton seed and similar oils"; and the specification also declares that the object of the patent is to recover more completely the more valuable constituents, of which the neutral oils are admittedly the chief. Other parts of the original specification refer to an example of a 12 per cent alkaline solution; to complete inter reaction between the alkali and every molecule or globule of oil; to an ultimate product that is edible, which could not be if rancid free fatty acids were retained in the refined oil; and to the removal from the oil by the centrifuge of the foots or soap stock resulting from the alkali treatment. Comparison is made between the patented process and the earlier and less efficient batch-kettle process which every one admits was used for the complete refining of neutral oils. In short, the original specification is devoted to a caustic soda refining process to neutralize the fatty acids and to produce soap stock, and this process usually precedes the reaction for the removal of color and gums.

In October, 1937, the specification was amended by adding a new drawing suggested by the Patent Examiner, of which no criticism is now made, and by adding three new paragraphs to the written specification which are objected to on the ground that they radically altered the original disclosure. The defendant says that they introduced new matter in the specification so that for the first time it was made to include statements in respect to "reacting on fatty acids with lye to form soap stock", "conditioning the mixture", "emulsion breaking temperature", "control of temperature to suit the oil being refined", and "varying the proportions of lye to accommodate the amount of fatty acid."

Of these enumerated particulars the most significant is the first named since it describes the fundamental character of the process; but as we have seen, it was not new to the patent in any sense for it was unmistakably set out in the original specification. The situation was practically the same with regard to the rest of the added matter, for the amendment was merely an elaboration, designed to explain and make more clear that which was already reasonably indicated by the original specification to be within the scope of the invention. The original specification obviously taught the liberation of the fatty acids from the oil by the use of such quantity of the reagent as would produce this result, having due regard to the quantity of oil to be refined, and also taught the application of heat to the mixture at a controlled temperature in order to break down the emulsion and facilitate the separation of the ingredients of the mixture in the centrifuge. The original specification, as we have seen, called for a coordination of the volume of input, output and the volume of heat to establish the proper velocity and turbulence of the oils within the coils; and it was stated that the exact procedure and reagent could be varied to suit the various oils depending upon their nature and the ultimate product desired without departing from the spirit of the invention. The evidence shows very clearly that any person skilled in the art would be able to select the appropriate strength of the caustic solution and the appropriate temperature to use in the refinement of any particular kind of oil. In like manner men of experience in the industry were accustomed to adapt the batch-kettle process to the changes in the crude oil submitted for refining.

The authorities support the propriety of the amendment to the specification made by Clayton in 1937 for "the rule is that insertions by way of amendment in the description or drawing, or both, of a patent application do not invalidate the patent, if they are only amplification and explanation of what was already reasonably indicated to be within the invention." Michigan Carton Co. v. Sutherland Paper Co., 6 Cir., 29 F.2d 179, 184.

It is significant that the leading refineries who took licenses from Clayton had no difficulty in recognizing the true nature of the new process as one designed primarily to refine neutral oil with the formation of soap stock; nor did they have any difficulty in proving the practical efficiency and superiority of the process over the method formerly in vogue. It is equally significant that when the patent specification and claims were considered from the technical viewpoint of the patent law, the Patent Office had no doubt as to the sufficiency of the original specification to support the claims. Indeed the very claims now sued on were awarded to the patent by the Patent Office before the specification was amended and after an adversary proceeding in which it was subjected to intensive scrutiny. Furthermore, Clayton was held to be entitled to claim 2 of the patent in suit before the amendment was made, and this conclusion was affirmed by the Court of Customs and Patent Appeals. James v. Clayton, 90 F.2d 337, 24 C.C.P.A., Patents, 1329.

Claim 2, which was before the Patent Office and the court during the interference proceeding, is as follows:

"2. In the purification of vegetable and animal oils containing free fatty acids, a continuous process comprising intimately mixing a measured quantity of oil with a measured quantity of alkali for a brief period, thereafter passing the mixture continuously through a heated conduit to raise its temperature to a degree which will facilitate centrifugal separation and then promptly centrifugally separating the refined oil from the resulting sludge and residual solution."

Claims 9, 10 and 15, which are typical of the remaining claims allowed by the Patent Office after the interference and before the amendment of the specification in 1937 are as follows:

"9. In the purification of vegetable and animal oils containing free fatty acid, a quick continuous process comprising mixing measured quantities of oil and alkali for a brief period to effect substantial neutralization of the free fatty acid contained in the oil and to form soap stock, thereafter advancing the mixture to a centrifugal separator while maintaining the same in a state of movement sufficient to cause the soap stock to be in such a degree of diffusion as to promote substantially uniform separation, promptly centrifugally separating the neutral oil from said soap stock, the process being characterized by limiting the time of contact of the alkali and oil to such a brief period as to minimize the saponification of the neutral oil and by the mixture of soap stock and oil having an emulsion breaking temperature at the time of undergoing centrifugal separation whereby to condition the mixture for separation of the soap stock from the oil."

"10. In the purification of vegetable and animal oils containing free fatty acid, a quick continuous process comprising the steps of pumping together relatively small metered streams of oil and alkali in substantially constant, predetermined proportions, mixing said streams for a brief period to effect substantial neutralization of the free fatty acid contained in the oil and to form soap stock while excluding substantial quantities of air, promptly advancing the mixture by pump pressure to a centrifugal separator for the prompt separation of the neutral oil from the soap stock, the process being characterized by limiting the time of contact of the alkali and oil to such a brief period as to minimize saponification of the neutral oil and by the mixture of soap stock and oil having an emulsion breaking temperature at the time of undergoing centrifugal separation whereby to condition the mixture for separation of the soap stock from the oil."

"15. In the purification of vegetable and animal oils containing free fatty acid and color impurities, a quick continuous process comprising mixing relatively small metered quantities of oil and alkali for a brief period to effect substantial neutralization of the free fatty acid contained in the oil and to form soap stock and conditioning the mixture for the step of centrifugal separation by the presence of an emulsion breaking temperature, advancing the thus conditioned mixture to a centrifugal separator under superatmospheric pressure and maintaining the mixture in a state of movement during the advancement thereof to prevent substantial stratification of the mixture while providing sufficient time for color reduction, promptly separating the neutral oil from the soap stock, the process being characterized by limiting the time of contact of the alkali and oil to such a brief period as to minimize saponification of the neutral oil and by the mixture of soap stock and oil having an emulsion breaking temperature at the time of undergoing centrifugal separation whereby to condition the mixture for separation of the soap stock from the oil."

With the same erroneous conception of the nature of the original specification in mind, the defendant contends that the patent in suit is invalid, because the claims were changed in important particulars five years after the defendant entered the field. On August 13, 1932, Clayton presented the interference counts 1 and 2, which included claim 2 of the patent now in suit. In May, 1933, the defendant adopted its Process A. On October 27, 1937, shortly after the termination of the interference proceeding, the patent application was amended by adding claims 9, 10, 11, 15, 22, 27 and 30 now in suit. In the late fall of 1938 the defendant adopted its Process B. Each claim in suit covers defendant's Process A, and these claims include claim 2, which, as we have seen, had been offered by Clayton before the defendant adopted its Process A. Claims 9, 10, 11, 15, 22 and 30, which applied to Defendant's Process B, were presented before the defendant adopted that process. Thus in neither case did the defendant's process begin before a claim was filed which dominated it and the defendant in no sense acquired an intervening right. In fact, the defendant was well aware of the nature of the plaintiff's process from and after the year 1932. It is true that Process A antedated claims 9, 10, 15, 22, 27 and 30; but claim 2 preceded and dominated Process A, so that the delay in presenting the subsequent claims in no way prejudiced the defendant. See Crown Cork & Seal Co. v. Ferdinand Gutmann Co., 304 U.S. 159, 58 S. Ct. 842, 82 L. Ed. 1265; General Talking Pictures Corp. v. Western Electric Co., 304 U.S. 175, 546, 58 S. Ct. 849, 82 L. Ed. 1273; Reo Motor Car Co. v. Gear Grinding Mach. Co., 6 Cir., 42 F.2d 965; Carson v. American Smelting & Refining Co., 9 Cir., 4 F.2d 463; Dwight & Lloyd Sintering Co. v. Greenawalt, D.C.S.D.N.Y., 20 F.2d 533; Wirebounds Patents Co. v. Saranac Automatic Mach. Corp., 6 Cir., 37 F.2d 830; Chapman v. Beede, 54 App.D.C. 209, 296 F. 956; Jenks v. Knight, 90 F.2d 654, 655, 24 C.C.P.A., Patents, 1227.

The Defense That the Claims Are Functional.

The defendant strives to show that the claims in suit are invalid on the ground that the description of the invention in the claims is merely functional. In General Electric Co. v. Wabash Appliance Corporation, 304 U.S. 364, 58 S. Ct. 899, 901, 82 L. Ed. 1402, the court condemned a claim which described a tungsten filament as composed "of comparatively large grains of such size and contour as to prevent substantial sagging and offsetting", which were the defects that the invention was designed to cure. A filament made up of comparatively large grains aptly described an earlier product and the added clause in respect to the structural character of the filament, apart from its statement of function, did not distinguish the patented product from the earlier filament. Nothing said in the specification in regard to the product or the process attempted to describe the filament except by mention of its coarse grained quality. The claim was therefore held invalid as an attempt to monopolize the product no matter how created. See, also, United Carbon Co. v. Binney & Smith Co., 317 U.S. 228, 63 S. Ct. 165, 166, 87 L.Ed. ___.

The defendant seeks to apply these decisions to the pending case by pointing out a lack of definiteness in the claims. For example, claim 2 describes a continuous refining process which includes passing the mixture through a heated conduit to raise its temperature to a degree which would facilitate centrifugal separation without specifying the precise degree of temperature; and claim 9 describes a quick continuous refining process which includes mixing measured quantities of oil and alkali "for a brief period to effect substantial neutralization of the free fatty acid contained in the oil and to form soap stock." It is contended that by reason of these omissions, the claims become merely descriptive of the desired function.

This argument, we think, misconceives the nature of the claims in suit. They do not state merely the results to be achieved but are process claims which set out a series of steps by which the desired ends are attained with sufficient definiteness to inform persons skilled in the art and to limit the extent of the monopoly of the patent. It was not necessary or practicable for Clayton to specify the exact degree of emulsion breaking temperature or the precise length of time needed for the substantial neutralization of every quality of oil that might be subjected to refinement; and the evidence shows that the information given was entirely adequate to persons skilled in the art. That such a description is sufficient is shown in Smith v. Snow, 294 U.S. 1, 7, 55 S. Ct. 279, 79 L. Ed. 721, wherein a claim relating to incubation of eggs was sustained although it spoke only in general terms of the size of openings for ventilation and the velocity and temperature of a current of air to be employed in the patented method. See, also, Waxham v. Smith, 294 U.S. 20, 55 S. Ct. 277, 79 L. Ed. 733; Gulf Smokeless Coal Co. v. Sutton, 4 Cir., 35 F.2d 433; Colgate-Palmolive-Peet Co. v. Lever Bros. Co., 7 Cir., 90 F.2d 178; Catalin Corporation of America v. Catalazuli Mfg. Co., Inc., 2 Cir., 79 F.2d 593.

There are many situations in the practice of the arts in which specific directions are properly omitted from the claims of patents because greater definition is either impracticable or is unnecessary to inform the art, and would serve only unduly to limit the scope of the invention or to invite evasion by those who desire wrongfully to misappropriate the substance of the invention.

With respect to a similar situation in the art of mineral separation, the Supreme Court said in Minerals Separation, Ltd., v. Hyde, 242 U.S. 261, 270, 271, 37 S. Ct. 82, 86, 61 L. Ed. 286:

"Equally untenable is the claim that the patent is invalid for the reason that the evidence shows that when different ores are treated preliminary tests must be made to determine the amount of oil and the extent of agitation necessary in order to obtain the best results. Such variation of treatment must be within the scope of the claims, and the certainty which the law requires in patents is not greater than is reasonable, having regard to their subject matter. The composition of ores varies infinitely each one presenting its special problem, and it is obviously impossible to specify in a patent the precise treatment which would be most successful and economical in each case. The process is one for dealing with a large class of substances and the range of treatment within the terms of the claims, while leaving something to the skill of persons applying the invention, is clearly sufficiently definite to guide those skilled in the art to its successful application, as the evidence abundantly shows. This satisfies the law."

Defense Based on Prior Patents and the Prior Art.

The chief reliance of the defendant consists of certain prior patents and certain prior practices in the oil refining art. Consideration of this defense is facilitated by the unusual circumstance that the most pertinent of the patents in question furnished the foundation for practical experiments in the art that were performed by experienced men and found to be wanting. The defendant's expert at the trial below selected the Hapgood and Mayno patent of 1923, No. 1,457,072, as the nearest to Defendant's Process B, and the Clark patent of 1929, No. 1,741,756 as nearest to Defendant's Process A.

The Hapgood and Mayno patent is for a process for refining vegetable oils, especially cotton seed oil. Filtered oil runs to a tank of 60,000 pounds capacity, equipped with a heating coil whereby the oil is heated to a temperature from 80 to 100° Fahrenheit. Lye is contained in a separate tank of sufficient capacity. Measured streams of oil and lye from these two tanks meet in a funnel and flow by gravity into a mixer where mixing takes place and foots or soap stock is formed. The whole mixture then flows into a centrifuge where it is continuously separated into foots and refined oil. The percentage of oil and lye corresponds to defendant's commercial Process B. The time consumed in the process is stated to be from 5 to 15 minutes.

The object of this patent is generally the same as the patent in suit and similarities in the two processes are apparent; but, as we shall see, the Hapgood patent was the basis of large scale experimental work conducted by men of experience, including Mayno, one of the patentees, which was abandoned as unsatisfactory. A distinction between the two patented processes resides in the fact that the Hapgood process involves a gravity feed of the streams of oil and caustic to the mixer controlled by hand-operated valves upon the storage tanks. The rate of the supply to the mixer and to the separator is dependent upon the levels of the liquids in the tanks and the uncertainties of manual operation, and is subject to other variations due to differences in the viscosity and temperature of the oil and the fouling of the supply lines which interfere with the measured and constant feed of the ingredients. The result is that constant measured quantities of the ingredients are not introduced into the mixer and subjected to another mixing, as is done with the metered streams pumped under pressure in the Clayton process. Furthermore, Hapgood teaches that a primary saponification of the fatty acids takes place in the mixer, and a more intimate mixture of the constituents takes place in the centrifuge with a resultant second saponification of the fatty acids followed by separation. This method is opposed to the Clayton process which shows that the conditioning of the mixture and the completion of the refining reaction takes place in the mixer before the centrifuge is reached.

The second Hapgood and Mayno patent of 1925, No. 1,560,084 provides for a preliminary operation to separate the oil from the gums and coloring matter which it contains. The subsequent refinement of the oil by the removal of the fatty acids is like the process described in the first Hapgood and Mayno patent.

We are aided in the study and comparison of the Hapgood and Mayno process with the Clayton process by long and practical tests to which the former was subjected before the Clayton invention was made, and our determination, therefore, is not based merely upon a verbal comparison of paper patents. Both have been subjected to the practical test of commercial experience from which the Hapgood process emerged as a failure and the Clayton process as a success. In 1921 the DeLaval Separator Company, a manufacturer of centrifugal separators, entered the oil refining field and between 1921 and 1925 induced a number of business concerns interested in refining vegetable oils to undertake experimental work on a commercial scale upon a process that was based upon the Hapgood and Mayno patent. Mayno himself participated. The equipment that was used was built in the DeLaval experimental department and after a fair trial, in various locations under favorable circumstances with expert and practical supervision the equipment was in each case removed, and so far as practicable, was returned to DeLaval. Finally, in 1925, the work was laid aside and DeLaval's interest was not revived until 1932 when the Clayton process was brought to its attention. Then it supplied mechanical centrifuges to be used in the Clayton process.

The experimental operations were conducted at the Guttenberg plant of the American Cotton Oil Company in 1921, first in connection with a batch process, and later in 1923 with a centrifuge installed by Mayno under the supervision of Clarence B. Cluff, who was then the manager of that company's manufacturing department and is now head of the patent department of the defendant and testified for the defendant in the pending case. A substantial quantity of oil, said to amount to 1,000,000 pounds, was refined, but Cluff rejected the process in 1923, and the equipment was consequently scrapped and this ended the work at Guttenberg. The defendant was offered the process but rejected it for the reason that it was not as efficient as its own method which included the batch-kettle process.

In 1923, after the Guttenberg experiment, Mayno and others employed by DeLaval installed equipment for experimental operation at the plant of the Southern Cotton Oil Company in Portsmouth, Virginia. At the end of the experiment the company declined to purchase the equipment and it was removed by DeLaval. In 1923, again Mayno, acting for DeLaval, installed apparatus in the refining plant of Simonin's Sons & Company near Philadelphia where a substantial quantity of oil was refined in experimental operations which extended from the fall of 1923 until late in 1924. They were not successful and the equipment was removed. The evidence indicates that all the DeLaval experiments failed because of lack of control over the flow of the mixture, which, as we have pointed out, characterized the Hapgood and Mayno process and distinguished it from Clayton's.

All of these experiments were considered at length and in detail by the District Judge who reached the conclusion that none of them developed a reliable method of refining oil which could be practiced with certainty and profit and hence they were abandoned, and that no one has subsequently practiced the process which DeLaval attempted to develop. During the trial the judge also observed the action of a reconstructed setup of the Portsmouth equipment presented by the defendant, and noted the distinction between the apparatus and that actually now used by the defendant at its Portsmouth plant. He also pointed out that there were fundamental differences between the two processes and observed the inability of the defendant's experts in charge of the reconstructed apparatus to place themselves in the same state of ignorance of the process of the continuous refining of oil which confronted Mayno and his associates in their unsuccessful efforts that had taken place seventeen years before the trial. This conclusion is strengthened by an interdepartmental report of the DeLaval experts to the DeLaval Company on August 3, 1932, when the Clayton invention had become generally known and a review of the previous DeLaval experimental work was made. The experts reported that "so far as our work at that time is concerned, it would be considered an abandoned experiment."

This history not only establishes the distinction between the Clayton patent and process, on the one hand, and the teachings of Hapgood and Mayno and of the prior art on the other; but it also disposes of the contention that otherwise might be made, that even if Hapgood and Mayno did not anticipate, it was at least so close to Clayton that the latter's improvements must have been obvious to any one skilled in the art. It is difficult to conceive a test more pertinent or more favorable to Hapgood and Mayno than those undertaken by DeLaval between 1921 and 1925, supported as they were by ample funds and conducted for long periods under the joint supervision of competent experts and successful men of affairs. We do not need to indulge in conjecture. We know that the Clayton process was beyond the skill of persons whose expert knowledge and practical experience were superior to those of the ordinary skilled worker in the art.

The Clark Patents.

The Clark patents, No. 1,553,141 of 1925 and No. 1,741,756 of 1929, relate to oil purification and to an oil purifying apparatus respectively for the purification of contaminated mineral oils which have been subjected to detrimental use. The Clark apparatus includes a pair of pumps, linked together, by which the oil and purifying reagent are drawn from separate tanks into the purifying apparatus which includes an agitating tank, a heating coil and a centrifuge separation. It is said that this mechanical set-up was sufficient to suggest the procedure of the Clayton method and to deprive it of the quality of invention. There is no showing that the patents have met with success. The process is different from Clayton's, since it relates to mineral oils that are not saponifiable and does not touch one of the major problems that Clayton solves. Moreover, in Clark the impurities are removed in solution rather than by precipitation as in the Clayton process. It is significant that the Clark patents were issued to the Sharples Company, which was also an assignee of the James process; but notwithstanding the actual possession of all that Clark taught, James, who conceived the invention in suit as early as May 14, 1930, was unable to reduce it to practice prior to Clayton's application on May 2, 1931. So the court found in the interference proceeding, and therefore awarded priority to Clayton. See, James v. Clayton, 90 F.2d 337, 339, 24 C.C.P.A., Patents, 1329. It was held by the District Court, and we agree, that no refiner has ever considered the Clark patents adaptable to the refining of vegetable oil. Their mere existence is insufficient to invalidate the Clayton patent and to deprive the inventor of the reward for his substantial contribution to the improvement in the art.

Infringement.

It was adjudged by the interlocutory decree of the District Court that all the claims of the patent in suit, to wit, claims 2, 9, 10, 11, 15, 22, 27 and 30 were infringed by defendant's Process A, and that claims 9, 10, 11, 15 and 22 of the patent were infringed by defendant's Process B. Process A is a continuous process of only three minutes duration. Predetermined constant preparations of oil and caustic soda are propelled through a mixing zone which includes a mixture meter and a mechanical mixer that promotes swift, intensive incremental mixing. The emulsion is then passed through heated conduits wherein it is broken down, and the particles of soap stock are agglomerated for subsequent effective separation, which takes place in the next step of the process when the stream reaches a centrifugal separator. Thus it appears, and the conclusion is supported by uncontroverted expert testimony, that Process A comprehends all the elements of the Clayton patent, including claim 2.

Process B requires six minutes for the complete refining reaction. The apparatus and the process differ from Process A only by the substitution of a conditioning chamber for a conditioning coil, and by the application of heat before rather than after the mixing of the ingredients. The District Court held and the evidence shows that the conditioning chamber is nothing more than a larger enclosure than the Clayton coil and that the movement through the chamber, although somewhat slower, is continuous. The chamber is merely the equivalent of the coil. The evidence also shows that it is of no moment whether the oil is heated before or after mixing, for in each instance the same results are obtained in substantially the same way. In short, Process B involves no substantial departure from the teachings of the patent. The infringement found by the District Judge is clearly sustained by the evidence.

Affirmed.