GILBERT, Circuit Judge.
This is an appeal from the decree of the court below sustaining letters patent No. 835,120, issued to Sul-man, Picard, and'Ballot, on November 6, 1906, and assigned to Minerals Separation, Limited, and Minerals Separation American Syndicate, Limited, adjudging the appellant herein to have infringed the same, and enjoining further infringement. The patent is for new and useful improvements in ore concentration. Its object is to separate metalliferous matter from gangue, by means' of oils and fatty acids which have a preferential affinity for metalliferous matter. In the specifications reference is made to United States Letters Patent No.. 777,273, granted to A. E. Cattermole. The Cattermole patent specifies that an amount of oil, varying from 4 per cent, to 6 per cent, of the weight of metalliferous-matter present, is agitated with an ore pulp, so as to form granules-which can be separated from the gangue. The specifications of the patent in suit state that the inventors have found that:
“If the proportion of oily substance be considerably reduced (say, to a fraction of 1 per cent, on tbe ore), granulation ceases to take place, and after vigorous agitation there is a tendency for a part of the oil-coated metalliferous matter to rise to the surface of the pulp in the form of a froth or scum.”
And the specifications add:
'“The proportion of mineral which floats in the form of froth varies considerably with different ores, and with different oily substances, and, before utilizing the facts above mentioned in the concentration of any particular ore, a simple preliminary test is necessary to determine which oily substance yields the proportion of froth or scum desired.”
There are 13 claims in the patent. The first is:
“The herein described process of concentrating ores, which consists in mixing the powdered ore "with water, adding a small proportion of an oily liquid having a preferential' affinity for metalliferous matter (amounting to a fraction of 1 per cent, on the ore), agitating the mixture until the oil-coated mineral matter forms into a froth, and separating the froth from the remainder by flotation.”
The next three claims specify the quantity of oil as amounting to “a fraction of 1 per cent, on the ore.” The second claim adds to the first the use of slightly acidified water. The third adds to the second [102]*102“warming the mixture.” The fourth adds to the third “and removing the oily coating from the mineral.” The fifth specifies the oil as “oleic acid of a quantity of from 0.02-0.5 per cent, on the ore.” The sixth adds to the fifth the use of water containing 1 per cent, of sulphuric acid. The seventh adds “warming the mixture to 30°-40° centigrade.” The eighth specifies the use of oleic-soap solution, to produce oleic acid, “amounting to 0.02-0.5 per cent.” The ninth is:
“The process of concentrating powdered ores which consists in separating the mineral from the gangue by coating the mineral with oil in water containing a small quantity of oil, agitating the mixture to form a froth, and separating the froth.”
The tenth adds to the ninth “warming the mixture”; and the eleventh adds the use of acid. The twelfth adds, “separating the froth from the remainder of the mixture”; and the thirteenth adds a current of water to carry off the coarser minerals, and “filtering off the froth and removing the oleic acid therefrom by treatment with an alkali.”
The answer of the appellant denied that Sulman, Picard, and Ballot were the first inventors of the process, or that there was any invention described in the patent, and denied infringement, and alleged that the process described in the patent was old and not patentable; that the process had been included in certain patents which were enumerated in the answer. The appellant was adjudged to have infringed claims numbered 1, 2, 3, 5, 6, 7, 9, 10, 11, and 12.
The appellees’ process depends primarily upon the affinity of oil for the metalliferous portion of powdered ore when mixed with water. It is conceded that that affinity and the fact that oil will carry the metal-liferous portions to the surface of the mixture while the rock or gan-gue will sink, have been known for many years.* That which is presented as new in the patent, and as the pivotal discovery on which its validity depends, is the formation of a froth or scum containing the metalliferous matter produced by agitation of the pulverized ore in water, by the action of oil in a quantity less than 1 per cent, of the quantity of ore treated.
Turning to the- patents which are adverted to as showing the prior art, we find the following: The British patent to Haynes, No. 488, issued in 1860, recommends the use of from 11 to 25 per cent, of oil by weight to the finely powdered ore; the mixture to be agitated with water, either warm or cold, until the earthy matter sinks and the metal is gathered by the oil. In that process the sunken earthy matter is removed, fresh ore is replaced, and the operation repeated with the same oil until it will take up no more metalliferous matter. The United States patent, No. 348,157, issued to Everson, in 1886, specifies the use of oil in quantities varying from 5 to 18 per cent. Miss Everson was the first to make the important discovery that the affinity of the oil for the metal was increased by the addition of an acid. ' In the description of her process, she states that, in the operation, the mass is broken up and thoroughly stirred in water, in a vessel provided with a mechanical stirrer, and having an outlet or outlets at the bottom for the escape of the water and sand. The Schwarz United States, pat[103]*103ent, No. 807,503, applied for in May, 1904, and granted in December, 1905, describes a process in which the dry ore is first mixed with enough oil to make a thick pasty mass, and water is thereafter added, and the mass is agitated so as to cause the metallic particles to float. The quantity of oil to be used is not further specified, but it is in the testimony that the Schwartz process has been experimentally used with cotton seed oil; the quantity used being 3.6 per cent, of the ore treated. The United States patent to Kirby, No. 809,959, was applied for December 14, 1903, and granted January 16, 1906. The oil, used by Kirby was kerosene oil in which 5 per cent, of bitumen was dissolved. One of the claims is as follows:
“The process of separating minerals, which consists in mixing together the pulverized mineral material, a considerable quantity of water and a solution of bitumen in a distillable hydrocarbon liquid, the proportion of bitumen in solution being substantially sufficient to insure the coating and entrainment of the mineral particles; and allowing the same to settle, and removing therefrom the floating layer of said solution and the mineral particles which have been coated thereby.”
In his specifications he says:.
“These materials to be so thoroughly agitated together as to finely subdivide said solution into small globules, and bring said globules into contact with substantially all of the pulverized mineral particles, which will by preference adhere to them.”
And, in describing the means for the agitation, he specified a “vertically rotating shaft” in the mixing tank, which it is said is to be “rotated rapidly.” No specific quantity of oil is prescribed, but it is stated in the specifications that a sufficient amount of the kerosene and bitumen in solution is.to be used; “excellent results being obtained by using one-fourth to three-fourths as much by weight as ore.”
In the Froment Italian patent, issued May 20, 1902, and the British patent to Froment, issued June 4, 1903, the invention described is declared to consist “of a modification of what is known as the oil process of ore concentration.” The modification consisted in releasing gas from the finely powdered ore, and “adding a suitable oil,” resulting in a.flotation of the oil and metal to the surface. Froment mentions the use of gas to aid in the flotation of sulphides reduced to powder and moistened by a fatty substance, in explanation of which the patentee says that if, for example, in a test tube there is placed ten grams of sulphurated copper ore, with its gangue, a gram of limestone, the whole reduced to powder, and there is added thereto 30 grams of water, a few drops of sulphuric acid “and a thin layer of ordinary oil,” and the mixture is agitated for a second, the whole of the copper pyrite will instantly rise to the top of the liquid. The patentees of the appellees’ patent purchased from Froment on November 17, 1903, the Froment patent, and Froment agreed to send all possible information, together with the plans, drawings, and a model plant, to England, for the use of the purchasers. The agreement was carried out in December, 1903. The instructions which Froment sent contain the following:
“If the ore contains more than 5 per cent, of metallic matter, such as copper, lead, it will be necessary to use a little more oil. As a general rule one [104]*104may assume l.per cent, of oil for ore containing up to 5 per cent, of metals, 1% Per cent, of oil for ore containing up to 10 per cent, and so on, up to ores containing 50 per cent of metallic lead, wMeh it was said would require 3% per pent, of oil.”
In the instructions the mixing device is described as composed of a cylindrical body made of strong sheet iron, riveted with bolts, “in which two stirring devices work in opposite directions, making about 300 revolutions per minute.”
The Cattermole process, which is referred to in the letters patent in controversy in this suit, is presented in two United States patents, one No. 777,273 and one No. 777,274, both of date December 13, 1904. The amount of oil specified in the Cattermole process is from 4 to 6 per cent. of the weight of metalliferous mineral matter present in the ore. This quantity with the average ores would be less than 1 per cent, of the mass treated. The fact that the final purpose of the Cattermole process is to increase the sinking tendency of the metalliferous mineral instead of removing it upon the surface does not render that process any less instructive as to the state of the prior art, for there are two ■distinct steps in the treatment of the ore; there is first a violent agitation to cause the oil to gather up the metalliferous particles, as in the case of other oil flotation processes. Their follows a slow, smooth stirring of the mass, which releases the air from the froth or scum and causes the oil and metalliferous particles to adhere together and form granules of sufficient size and weight to sink, after which the mass at the bottom is drawn off, and the particles of sand and gangue therein are forced upward and thus separated from the granules of metal. Here is described a process in which the quantity of oil used approximates the quantity which is called for by the appellees’ patent, a fraction of 1 per cent.
. When the claims and the description of the process of the appellees’ patent are compared with the patents of the prior art, it will be seen that the only material difference is in the smaller quantity of oil which, the appellees use. Comparing the appellees’ process with the Froment process, it will be seen that in both powdered ore is mixed with a sufficient water to form a freely ■ flowing pulp; that while Froment recommends the use of oil of from 1 per cent, to 3% per cent, of the ore by weight, the appellees recommend a fraction of 1 per cent, of oil; that Froment recommends the use of ordinary oil or a suitable oil, while the appellees’ patent mentions oils, fatty acids, and oleic acid; that in both processes there is agitation of the mixture; that in both processes a. small quantity of sulphuric acid is recommended; that in both processes the mixture of all the substances is made before the agitation; and that in both processes bubbles of air or gas are caught in the sulphides in the form of a froth which rises to the surface. So also, in the Kirby patent, the various steps of the process are similarly described. We find there the pulverization, of the ore, mixing it with water to form a floating pulp, mixing therewith kerosene oil with a small percentage of bitumen, violently agitating the mass so as to bring the oil into contact with the mineral particles of the ore, causing the oil-coated mineral particles to rise and float on the surface of the wa[105]*105ter, while the gangue sinks to the bottom, and lastly skimming the floating concentrate from the surface of the liquid.
[1] But the appellees say that the Froment patent is. a paper patent, and that therefore it is to be disregarded. A paper patent, if it fully describes an invention, whether it be a machine, device, or process, is just as effective to show anticipation as a patent which describes an invention which has gone into extensive use, for a presumption of operativeness and of some utility attends the granting of letters patent. Packard v. Lacing Stud Co., 70 Fed. 66, 16 C. C. A. 639; E. L. Watrous Mfg. Co. v. American Hardware Mfg. Co. (C. C.) 161 Fed. 362; National Chemical & Fertilizer Co. v. Swift & Co., 104 Fed. 87, 91, 43 C. C. A. 421; Universal Winding Co. v. Willimantic Linen Co., 82 Fed. 228; Van Epps v. United Box Board & Paper Co., 143 Fed. 869, 75 C. C. A. 77; Ironclad Mfg. Co. v. Dairyman’s Mfg. Co., 143 Fed. 512, 515, 74 C. C. A. 372; Dashiell v. Grosvenor, 162 U. S. 425, 432, 16 Sup. Ct. 805, 40 L. Ed. 1025. In Telephone Cases, 126 U. S. 1, 536, 8 Sup. Ct. 778, 783 (31 L. Ed. 863), it was said:
“The law does not require that a discoverer or inventor, in order to get a patent for a process, must have succeeded in bringing his art to the highest degree of perfection. It is enough if he describes his method with sufficient clearness and precision to enable those skilled in the matter to understand what the process is, and if. he points out some practicable way of putting it in to operation.”
[2] The fact that the appellees use a smaller quantity, of oil than was used in the prior art is not of itself, and it is not claimed by them to be, sufficient to distinguish their process so as to render it patentable. To discover that the desired result may be accomplished with the use of a fraction of 1 per cent, of oil when formerly a much larger quantity of oil had been used, and had been deemed necessary, is not an invention or discovery within the meaning of thé patent laws. It is a difference of degree and not of kind.
“ ‘A change only in form, proportions, or degree, doing substantially the same thing in the same way, by substantially the same means, with better results,’ * * * is not such invention as will sustain a patent.” Roberts v. Ryer, 91 U. S. 150, 23 L. Ed. 267.
In Fried. Krupp Aktien-Gesellschaft v. Midvale Steel Co., 191 Fed. 588, 112 C. C. A. 194, the court said:
“But mere useful and economical administrative methods, however valuable, while they may and usually are incident to invention, do not themselves constitute invention.”
And in De Lamar v. De Lamar Min. Co., Ltd., 117 Fed. 240, 249, 54 C. C. A. 272, 281, this court, referring to the use of zinc dust for the purpose of precipitating mineral in a solution, said:
“The right to use the dust being free to all, we think it follows, necessarily, that all have the right to adjust the quantity of the material to the necessities of each case, and to ascertain by experiment or analysis, if need be, the quantity that may be required to produce the desired end, and that such a use cannot be made the subject of monopoly; there being involved in it no-discovery, but only the exercise of ordinary prudence and skill.”
See, also, Brady Brass Co. v. Ajax Metal Co., 160 Fed. 84, 87 C. C. A. 240; Commercial Mfg. Co. v. Fairbank Co., 135 U. S. 176, 10 [106]*106Sup. Ct. 718, 34 L. Ed. 88; Bullock Electric Mfg. Co. v. General Electric Co., 149 Fed. 409, 79 C. C. A. 229; Lauman v. Urschel White Lime Co., 136 Fed. 190, 69 C. C. A. 206; Tilghman v. Proctor, 102 U. S. 707, 26 L. Ed. 279; Guidet v. Brooklyn, 105 U. S. 550, 26 L. Ed. 1106.
But it is said that the appellees’ process is the only froth process, and that the processes described in the prior art are the bulk'flotation or oil buoyancy process. We have to inquire, therefore: What is the oil buoyancy process. It is described in the Elmore patents, No. 676,-679,'issued June 18, 1901, and No. 689,070, issued December 17, 1901, and in the testimony of Mr. Ballantyne, general patent counsel of the appellees. Mr. Ballantyne said that it was essential to this process: First, that the oil should be thick and viscous, for otherwise the metal would easily fall out of the oil; second, that the process be carried out in the cold, because, if the oil be heated, it becomes thin and will not entrap or float the mineral; third, that the mingling of the pulp and the oil be as gentle as possible, so to avoid breaking the oil into globules. And the appellees’ expert, Dr. Chandler, testified that, in the oil bulk flotation, 15.7 parts of oil must be used to 6.7 parts of ore. With these definitions and explanations of the oil buoyancy process in view, it is very plainly to be seen that in none of the patents herein discussed, aside from the Elmores’, is that process used. The Haynes patent uses 11 per cent, of oil to the ore, and directs that the mixture be agitated with hot, warm, or cold water. Kirby recommends thorough agitation by rapid rotation of the agitator, and he refers to the floating concentrate as a “scum.” The Everson patent declares that thorough agitation of the mass or pulp, comprising water, the finely divided ore, the oil of fat, and the acid is necessary, and in the Engineering and Mining Journal of November 15, 1890, there is an Article describing a test operation of that process, in which it was said that as the result thereof “a thick scum of sulphurets rose to the surface and was skimmed off, leaving the hitherto black ore as white as snow; in fact, pure silica.” And so in the Cattermole process, although its ultimate purpose was to precipitate the metal in the 'form of granules to the bottom of the tank, it was first required that the mixture be thoroughly agitated, and, as the result thereof, a froth necessarily came to the surface.
The contention is made, however, that if, indeed, the prior patents disclose froth processes, the scum which the appellees’ process causes to rise to the surface of the water is so different from the froth or scum which arises under the prior art that it is a new result and is not anticipated by anything in the prior art. Say counsel for the appellees, “You produce a froth which is not our froth;” and again they say that the appellees’ froth is “a pure metal froth,” while that of the other patents is an “oil froth,” and that the appellees’ froth consists of air bubbles surrounded by metal armor without any appreciable quantity of oil connected therewith. The evidence in the case, together with the illustration thereof afforded by demonstrations of the various processes which were made in the aid of the argument before this court, convince us that the froth in all these processes is the same, [107]*107with the exception that there is less oil (as there must necessarily be) in the appellees’ froth than in the others. The froths áre all similar in appearance; they all rise to the surface after the same amount of agitation ; they all gather with equal efficiency the same quantity of metal ; and all may be removed from the surface in the same way. It is not literally true, of course, that there is no oil in the froth which the appellees produce. If there were no oil in it, there would be no froth. The quantity of oil that has been put into the mixture must necessarily be found in the froth. It would manifestly be a physical impossibility to create a froth consisting only of bubbles of air and the particles of mineral. This is admitted in the specifications of the appellees’ patent, in which it is said that there must be agitation “until the oleic acid has been brought into efficient contact with all the mineral particles in the pulp,” and that the air bubbles which make the scum floatable adhere “only to the mineral particles which are coated with the oleic acid.”
It is said in the appellees’ brief that one feature of the froth produced by their process is that it is unnecessary to clean the mineral of the oil, as the infinitesimal oil coating has no adhesive effect, and that the concentrates may be freely tabled or vanned; and counsel say, “All you have to do is to gather that froth from these vessels, dry it out, and take it to the smelter.” We find no; evidence in the record that the froth which the appellees produce may be smelted without removing the oil therefrom, or that it would be a distinct advantage to do so, or that the froth produced by some of the anticipating processes may not be taken to the smelter without removing the oil. If the statement made in the brief is true, it is very clear that that feature of their froth was not known to the patentees at the time of their application for the patent, and is nowhere disclosed in the specifications or claims. The specifications deal only with the production of the scum or froth; but claim 13 includes “filtering off the froth añd removing the oleic acid therefrom by treatment with an alkali”; and on May 3, 1905, in a statement of their method of oil concentration which they sent to Australia; the patentees said:
“Tlie oleie acid is therefore of necessity recovered in the process and in the state in which it is required for re-use. This is not undertaken merely for the purpose of recovering the oleic acid, but as a necessary step in rendering the concentrates fit for vanning, separation into blende and galena.”
[3] The decision of'the court below appears to have been largely influenced by the consideration that the appellees’ patent had gone into extensive and successful use. The fact that a patented device or process has gone into extensive and successful use is often of value in determining the question of invention and patentability. It is referred to for the purpose of turning the scales in cases of grave doubt. It is of no value whatever where the question of the invention or patenta-bility is free from doubt, and in any case its value depends largely upon .the causes which produced it. It is often due to business ability in .manufacturing, exploiting, and advertising, and to the fact that prior conditions have not stimulated development. The appellees’ process, originally patented in Great Britain, has been installed in Australia, Sweden, Finland, Chile, and Wales, and it is in the process of installa[108]*108tion in Cuba. It is not improbable that in those'countries the prior art may have been substantially unknown, and it is possible that the ap-pellees’ success there is referable to the fact alleged in the bill:
“That the complainants have been to great trouble and expense in the introduction into use of the process, and have invested large sums of money in its introdi]ction into commercial use in different parts of the world, and in the effort to introduce it into commercial use in the United States.”
It is in evidence that, in making the process known to the public in the United States, the appellees have expended $60,187. Notwithstanding these efforts and expenditures, their process has not yet been installed in the United States.
In Olin v. Timken, 155 U. S. 141, 155, 15 Sup. Ct. 49, 55 (39 L. Ed. 100), it was said:
“While the patented article may have been popular and met with large sales, that fact is not important when the invention is without patentable novelty.”
In McClain v. Ortmayer, 141 U. S. 419, 428, 12 Sup. Ct. 76, 79 (35 L. Ed. 800), the court said:
“That the extent to which a patented device has gone into use is an unsafe criterion even of its actual utility is evident from the fact that the general introduction of manufactured articles is as often effected by extensive and judicious advertising, activity in putting the goods upon the market, and large commissions to dealers, as by the intrinsic merit of the articles themselves.”
It would serve no useful purpose to revielv all the numerous decisions in patent cases that are cited by the appellees. But we will refer to two of them which seem to be especially relied upon, Naylor et al. v. Alsop Process Co., 168 Fed. 911, 94 C. C. A. 315, and United States Mitis Co. v. Midvale Steel Co. (C. C.) 135 Fed. 103. In the first of these cases the court held that an expert cannot take a process patent which has never been applied industrially and work the process in his labratory, and discover therefrom something which is not disclosed on the face of the patent, and then transfer that experience back to the time of the patent, and make it a part of the prior art for the purpose of defeating a subsequent patent for a meritorious invention. There is nothing in the present case to which that ruling applies, and it is to be observed that the court sustained the patent in that case expressly upon the ground that the complainant therein was the first to discover a successful- process for bleaching flour. The court said:
“His act was not selection of known agents in tbe art of bleaching .flour, but was the discovery of the only agent that has yet been found to accomplish that result successfully in the milling industry.”
The second case is said by counsel for the appellees to be “conclusive.” The invention there under consideration was a process for making steel castings homogeneous so as to prevent blow holes, and it -consisted in adding to the molten steel, when it is about to be poured into the mold, a minute quantity of metallic aluminum. But the distinction between that case and the case at bar is plainly to be seen in the fact that there the inventor made a distinct addition to the known .art. He added something which had not been used by others, and [109]*109which was effective in producing the result which was sought. In the present case nothing has been added.
[4] We hold that to sustain the appellees’ patent would be to give to the owners thereof a monopoly of that which others had discovered. What they claim to be the new and useful feature of their invention, as stated by their counsel, is “agitating the mixture to cause the oily coated mineral to form a froth.” As we have seen, that feature was clearly anticipated by the prior art, and, when the elements of the ap-pellees’ claims are read one by one, it will be found that each step in their process is fully described in more than one of the patents of the prior art, with the single exception of the reduced quantity of oil which they use. The patentees of the appellees’ patent made a valuable contribution to the art in discovering the smallest quantity of oil which would produce the desired result. In doing so, they pursued the course which all skillful metallurgists would be expected to pursue. They made a series of experiments to determine how small a quantity of oil could be used successfully. They found, as all must find who apply the oil flotation process, that certain oils are adapted to use with certain ores, and that a larger quantity of oil is necessary for one kind of an ore than for another. The appellees admit that for some ores they use four times as much oil as for others. Their discovery that a small fraction of .1 per cent, of oil is sufficient to produce flotation of the metalliferous matter cannot, as we have seen, be made by itself or in a combination the subject of a patent. The appellees cannot take from others the right to use oil economically. This was evidently the ruling of the Patent Office on their application for a patent. One of their claims in the original application was “the process of concentrating powdered ore, which consists in separating minerals from gangue by coating the minerals with oil in water containing a fraction of 1 per cent, of oil on the ore, and recovering the oil coated minerals.” This was rejected in view of the Cattermole patent “as expressing merely a difference of degree thereover as to the proportion of oily matter employed.” Counsel for appellees admit that the claim was properly rejected for the reason that it leaves out the agitation and froth, and say, “Our invention is something else than the mere reduction of oil.”
The decree is reversed, and the cause is remanded, with instructions to dismiss the bill.