That the process of the patent and the apparatus devised for the employment of the process accomplishes this result better and cheaper than any method or apparatus before known or used is abundantly established by the evidence. The evidence shows that the plant constructed and operated by the Tennessee Cotton Fiber Company of Memphis, Tenn., is successfully handling 24 tons of hulls per day, and is deriving, approximately, from this daily consumption 8 tons of fiber and 16 tons of bran. That Johnson was not the first to realize the value of these by-products of the cotton seed must be conceded. That he was not the first to make a bran partially free from fiber, and a fiber comparatively free from bran, must be also conceded. That he was the first to discover a process and devise an apparatus by which these cheap products could be produced in the best condition of each and at such low cost as to be commercially valuable is clearly shown by the evidence in the transcript. There remains therefore only the question as to whether his process and apparatus or either are novel and patentable.

First as to the process. Prior to Johnson's there were several methods for eliminating the lint from cotton hulls. One of the best known and most used was that of destroying the hulls by chemical agents which did not injure the cotton fiber. This involved a sacrifice of more than half the value of the raw material, and made the cost of the method altogether too great for commercial purposes. There can, of course, be no anticipation of Johnson's method, which was purely mechanical, by one which involved chemical agencies alone. Another method somewhat well known was that best shown by the patents to Emil Bohn of Galveston, Tex., being patents No. 437,084, issued September 23, 1890, and No. 438,984, issued October 21, 1890; the first being for a method, and the second for apparatus. Bohn's plan was to comminute the hull, including the fiber, force this commingled mass of hull and fiber through a screen for the purpose of separating the coarser parts of the hull with their still adhering fiber, and a recovery of such fiber as passed through the screen or separator by an air blast. The comminution of the hulls and fiber was accomplished by a grinding cylinder armed with a series of “sharp knife-like ribs," operating against a "series of knives" in the incasing cylinder. Bohn, in his process specification, says:

"In carrying out the invention the cotton-seed hulls and adherent fiber are introduced into a mill, A, and ground until the hulls and fiber are practically separated and reduced to comparatively small particles."

This mass of comminuted hulls and fiber is then carried in a confined chute to a screen, through which the finer particles are forced and then acted upon by an air blast; the theory being that the fine particles of fiber will be carried further by the air suction, the particles of hull as the heavier particles dropping by gravity into the first receptacle, while the lighter will be floated on to the second. The coarser particles which were not forced through the screen constitute what Bohn called "a marketable article of second grade." The material deposited by gravity in his first receptacle consists, says the patent, "principally of hulls and constitutes an article of food for stock." That which reached his second receptacle he says "consists of very fine particles of hulls and fiber and constitute the first-grade article." Bohn's

method and apparatus were employed by W. C. Johnson before he made his own discovery under the personal direction of Bohn. It was not a successful or commercially valuable experiment. The particles not passing through his screen, because of their coarseness, contained an excess of fiber, which, as fiber, was wasted and was injurious to the bran as a cattle food. His second-grade stuff was of little value as paper stock because it contained too much bran, and because the fiber was so comminuted by the grinding process as to be of less value as a paper stock than if saved as a larger fiber. From a ton of seed his average product was about 200 pounds of his "first-grade" fiber, and this, on account of the shortness of the fiber, was of diminished value.

The method of the patents to Burnett, Sears & Burnham of 1888 and 1890, being patents Nos. 380,087 and 440,259, is somewhat analogous to that of Bohn and encountered the same difficulties. Cotton seed or cotton-seed hulls were fed into a perforated cylinder in which a shaft revolved at high speed, armed with wings or beaters. The material is driven by these beaters with great force against the perforated walls of the cylinder by the centrifugal force of the rapidly rotating beaters, and the fiber, eliminated by the friction of the hulls against the rough walls of the cylinder and against each other, is forced through the meshes of the cylinder, together with the finer parts of the hulls. The particles of hulls are supposed to fall by gravity to the bottom of the chamber while the particles of fiber are floated by an air blast to another place of deposit. The method upon principle does not differ materially from that of Bohn. There is the same comminution of hulls and fiber. This mass is forced through the meshes of a perforated cylinder, which answers to the screen of Bohn, and a separation of those particles which get through by gravity; the lighter particles of fiber being carried farthest by a current of air. This method was used under remarkably favorable circumstances by the American Stock Food & Fiber Company at St. Louis. The product was subject to the same objections as that produced by Bohn. Neither constituent was separated in its best condition.

The plan was abandoned for another. That substituted for it was the method of the patent of J. P. Burnham of July 31, 1894, No. 431,391. This patent recognizes upon its face that the utility of the cotton fiber from cotton hulls is largely destroyed if the already short fibers of the lint are further shortened or ground up in the process of recovering them, and that the value of the hulls is also diminished if they are too finely ground up and the cotton lint not cleanly removed. To accomplish the separation without injury to either, Burnham avoided. all grinding of hull or fiber and undertook to break or eliminate the lint from the hull by centrifugal action. It did not prove a commercial success, for the St. Louis concern soon ceased to do business and sold its plant to others, who did nothing with it. Under the Burnham process the hulls were not converted into bran, or, if so, by an independent and subsequent operation, for under his plan the lint was to be separated from the unground or unbroken hulls by a process of attrition which was not intended to break them up at all. The method of Burnham, whether workable or not, was not the process of Johnson, and his patent

141 F.-6

is not an anticipating patent for that reason as well as because it had not been issued at the date of the Johnson application. Bates v. Coe, 98 U. S. 31, 25 L. Ed. 68; Dubois v. Kirk, 158 U. S. 64, 15 Sup. Ct. 729, 39 L. Ed. 895; Anderson v. Collins, 122 Fed. 451, 58 C. C. A. 669. The specifications of the Johnson patent betray a knowledge of the efforts theretofore made to produce a branless fiber and a fiberless bran from cotton seed and cotton hulls by mechanical means, and describes a process which he claims will accomplish the desired end by separating the two independent products in their best condition. The discovery he says consists "in breaking the hulls by attrition and concussion and then conducting the broken mass forcibly and rapidly into a separator which thoroughly separates the fiber from the hull and kernel." The mechanism suggested in which to employ his method, and the subject of the second or mechanical claim of his patent is shown below:

His description of his method is best understood when read in connection with the figure above set out. The patentee says:

"In carrying out my invention, I employ a rotary breaker or grinder, A, consisting of the disks, a, arranged vertically in close proximity and revolved rapidly in opposite directions by means of the shafts, a1, a, carrying the pulley wheels, a2, a2. The seeds and hulls with the adhering fiber are fed to the breaker or grinder by means of a hopper, a3. The disks are so constructed and arranged that the hulls and kernels are broken or split without cutting the fiber, and the revolutions of the two disks are such that the broken mass is

forcibly ejected Into a separator, B, located directly beneath the breaker or grinder. This separator, B, consists essentially of a U-shaped trough, b, formed of some strong bolting or screening material, such as wire gauze, the mesh thereof being very fine. A rotary brush, B1, is journaled within the said trough, said brush consisting of a series of oblique arms, b1, carrying brushes, b2, which contact with the face of the wire gauze or screen. As the broken hulls and kernels and fiber are received into the separator, the rapidly revolving brush forces them into contact with the wire gauze and carrying them around separates the fiber from the hull and forces said hulls and kernels through the mesh; and the fiber being restrained from passing through, is fed out at the end of the separator into a chute, C, by means of which it is led to a second breaker or grinder, D, similar to the breaker, A, except that the disks are arranged somewhat closer than in the first instance. The object of this second breaker is to break or grind any particles too large to be forced through the separator, B, and also to accomplish a further separation between the hulls and fiber. The hulls, etc., after passing through the second breaker, D, are received into a second separator, E, which is also similar to the one before described. The separated hulls, etc., which were forced through the first separator are received into a chute, C1, and by that means conducted to the second separator, where they are given another treatment, so that every particle of fiber may be separated and a uniform grade of food stock produced. The finished product is then conducted to a bag or other suitable receptacle by means of a chute, C2. The fiber has, by these various operations, been thoroughly separated from the hulls and kernels, and the operations have been such that the fiber has not been cut or injured, so that a uniform quality of long stock is obtainable which can be used in the manufacture of paper if so desired. The batting or wadding thus produced can be conducted directly from the second separator; but if a particularly fine quality of stock is desired, we lead it into a beater, F, consisting of a chamber, f, and a rotary beater, f, the revolutions of which are so rapid that the fiber is thrown upward toward a passage, f2, and all dirt and impurities will fall of their own weight to the bottom of the beater. Connected with the passage, f, is a conductor which leads to the collecting room."

The two claims in issue are as follows:

“(1) The process herein described, which consists in first breaking the hulls and seeds by attrition or concussion and forcibly ejecting the mass into a separator, conducting the fiber to a second breaker and subjecting the same to a finer and closer grinding operation, and then forcibly ejecting the ground mass into a second separator, the bolted mass from the first separator being led also to this second separator and finally conducting the separated fiber away from the second separator, substantially as shown and described.

"(2) In an apparatus of the class described, the combination with the vertical rotary disks, a, a, of the horizontal separator, B, arranged beneath the same, the chute, C, leading from the end of separator, the chute, C1, leading from the bottom of the separator, the second breaker, D, the disks of which are arranged closer together than the disk, a, of the first breaker, the horizontal separator, B, and the chutes for conducting the fiber and bolted material, substantially as shown and described."

The material to be dealt with was of a most difficult character. The fine, fuzzy fiber adheres very closely and stubbornly to the woody capsule. Johnson realized from his experience with the Bohn method, and his knowledge as a practical paper maker, the necessity of making the separation between the two constituents as clean as possible. As we have already seen under the methods of the old art, as illustrated by the Bohn patent and that to Burnett, Sears & Burnham, the hulls were subjected to grinding, whereby both hulls and fiber were ground into fine particles, so fine as to pass through the fine meshes of a screen or separator. The fiber was expected to pass with the bran through the

screen; a separation more or less thorough being made after so passing by an air blast. This method resulted in eliminating some of the fiber, but in such a comminuted condition and so commingled with the bran as to be of diminished value. Johnson's discovery was that, if the hulls were subjected to attrition as distinguished from grinding, two things would result: First, that more of the fiber would be cut loose from its attachment than by the other plan; and, second, that the detached fiber was separated in better condition, not being comminuted as by a technical grinding action. This being the condition of the material after a treatment by attrition, his next step was to separate the fiber so far as detached from the broken hulls with which it was commingled by means of a screening operation, whereby the fine bran is forced through the meshes of the separator. This fiber, by reason of its noncomminuted condition, he found did not pass with the fine bran through the meshes of his separator. This fiber, thus restrained from passing together with the particles of hull left in the separator, he subjects to a second treatment by attrition in a mill like the first, except that the disks are arranged closer together, the better to break the partly reduced hulls too large to pass the meshes of the separator and to further carry on the delinting process. The product of this second treatment, together with the material which passed through the meshes of the first separator, he carries to a second separator like the first; this second treatment by screening effecting a more complete separation of the bran and fiber. The success of the whole method is thus made to turn upon successive operations which avoid cutting the fiber while under treat

ment.

"The disks are to be so constructed and arranged that the hulls and kernels are broken or split without cutting the fiber."

The "fiber is restrained from passing through" the screen and is "fed out at the end of the separator.' The means, method, or process constitutes a patentable process, consisting, as it does, of a series of treatments of a peculiarly obstinate material, each treatment having relation to the character of the material acted upon and to the condition produced by the preceding treatment. The result is the separation of the material into its two distinct constituents; the bran being substantially free from fiber and the fiber practically free from bran. Mowry v. Whitney, 14 Wall. 620, 20 L. Ed. 860; Cochrane v. Deener, 94 U. S. 780, 787, 24 L. Ed. 139; Carnegie Steel Co. v. Cambria Iron Co., 185 U. S. 403, 22 Sup. Ct. 698, 46 L. Ed. 968; Am. Fibre Co. v. Buckskin Fibre Co., 72 Fed. 508, 18 C. C. A. 662. In Mowry v. Whitney, cited above, a patent was sustained for a process for making cast iron wheels, by retarding their cooling by a second application of heat, until all parts of the wheel were raised to the same temperature and then permitting the heat to subside gradually. In Cochrane v. Deener, cited above, the patent was for a process of manufacturing flour, which consisted in passing the ground meal through a series of bolting reels composed of cloth of progressively finer meshes which passed the very fine flour and retarded. the passage of certain impurities. It was held to be a patentable process. Defining a process, Mr. Justice Bradley in that case said: