(after stating the facts as above). The process patent illustrates and describes an apparatus for grinding logs into pulp. The specifications explain that the inventor’s object is to obtain pulp which shall have longer, stronger, and more uniform fibers than have been secured by former grinding processes, and that this object is to be attained by making suitable grooves or depressions- in the face of the grindstone, in which abraded fibers will lodge and thus obtain protection from further grinding against the undersurface of the logs as the stone moves on. Between these grooves is the grinding surface of the stone, which “is roughened in the usual manner,”
The claims are four in number, the first being addressed to the process and the others to the apparatus. It will suffice to quote claim 1:
“1. The process of reducing wood to pulp, which consists in subjecting the wood to successive abrasive actions in direction substantially perpendicular to the length of the fiber, whereby the fibers are separated out from the wood, and protecting said fibers that have been separated by preceding abrasive actions from being broken and injured during succeeding abrasive actions, and then removing said separated fibers.”
Coneededly, the sole element of novelty in this process, or in the apparatus by which it is applied, is the idea of protecting the fibers by means of grooves or pockets in the face of the grindstone. Plaintiff’s expert Williams testified that, “if you eliminate the grooves c, you have a stone dressed according to the old practice.”
With respect to the dimensions of these grooves, the specifications contain the following (page 2, lines 50-70):
“I have shown the.grooves c of triangular shape and disposed in a straight line across the surface of the stone, and parallel with its axis, I prefer to make the grooves one-eighth of an inch deep and one-eighth of an inch wide at their mouths, and to dispose them around the surface of the stone about three-quarters of an inch apart, measuring from center to center of the grooves. * * * But without departing from my invention the shape and dimensions of the grooves, the distance between them and the direction in which they extend across the stone may, of course, be varied.”
So far as this record shows, neither plaintiff nor any one else has ever successfully applied this preferred form of marking to a pulp stone. The circular of the Hall Process Corporation (Defendant’s Exhibit B) shows a spiral burr, designated as style No. 126A, with teeth three-quarters of an inch apart. Walker testifies that such a burr was used in 1915, when he .was with the Fitzdale Paper Company, in an attempt to demonstrate the so-called Hall process for grinding pulp; that the demonstration was made by Edward Hall and two others from the Hall Process Corporation, and was admittedly unsuccessful. Walker further says that in 1918 he disclosed to Mr. Sherman, of the Hall Process Corporation, the method which he had employed at Fitzdale, namely, a four to the inch straight-cut burr, making grooves not more than one thirty-second of an inch deep and a 12 or 14 point diamond burr to sharpen the ridges between the grooves. In return Walker received 100 shares of stock in the Hall Process Corporation, whose name was then changed to Hall, Ward & Walker, and Walker became a demonstrator for the company. He never demonstrated anything but the shallow marking during the year he was connected with the plaintiff corporation under its former name.
There is no contradiction of Walker’s testimony. Grinding with a stone thus marked apparently came to be known as the Hall process. It is clear that the inventor’s preferred form of operation, in which the stone is dressed with grooves one-eighth of an inch deep and three-fourths of an inch apart, has never been adopted in commercial practice. The accepted practice of dressing stones in the manner which plaintiff claims as the Hall process is to make grooves about one thirty-second of an inch deep, spaced not less than four to the inch, and to sharpen the ribs between them with a fine diamond burr.
The plaintiff, relying upon Hildreth v. Mastoras, 257 U. S. 27, 34, 42 S. Ct. 20, 66 L. Ed. 112, argues that a patentee is entitled to profit by experience, and to modify his preference, so long as he holds to his prinei*907pie, and that the Hall process patent taught the principle of collecting and protecting the abraded fibers in depressed grooves, even though the inventor had a mistaken conception of the proper dimensions of such grooves for the most perfect operation of his invention. Hence plaintiff contends that the accepted practice is within its patented process, while the defendant asserts that the patented process has never been practiced by any one.
Assuming that the patent may be construed to cover the accepted practice as above described, we are brought to a consideration of the prior art. But here we are met by plaintiff’s objection that the validity of the patents cannot be' questioned because the defendant is estopped by its license. Such estoppel, however, does not extend beyond the term of the license. Tibbe & Son Mfg. Co. v. Heineken, 37 F. 686 (C. C. S. D. N. Y.); Bituminous Products Co. v. Headley Good Roads Co., 2 F.(2d) 83, 86 (D. C. Del.); Eskimo Pie Corp. v. Nat. Ice Cream Co., 20 F.(2d) 1003, 1005 (D. C. W. D. Ky.); Tate v. B. & O. R. R. Co., 229 F. 141 (C. C. A. 4). This came to an end June 8, 1922.
As to any infringement or threatened infringement thereafter, the former licensee may avail of the same defenses as a stranger. The permanent injunction which was granted rests upon the threat of a continuing infringement, and the defendant was entitled to plead invalidity against the claim for injunctive relief. See Eskimo Pie Corp. v. Nat. Ice Cream Co., supra. As to the effect of estoppel upon the claim for damages, more will be said later.
Validity of the patent is attacked on the ground that the patent scheme, if construed to cover the accepted practice of shallow grooves, does not protect against regrinding, and that such protection would, indeed, be undesirable. The testimony is somewhat conflicting on this subject. Walker testified that the shallow grooves do not in fact furnish protection, and that regrinding the fiber beneath succeeding logs is essential, and, if not done on the stone, must be done with other refining instruments. Potter, who has spent his life in pulp making, also asserts that “regrinding is a benefit to the stoek.” Sheffield says that the purpose of the grooves is to carry water to prevent the grinding surface from overheating. Defendant’s witness Thickens, however, testified that shallow grooves do prevent regrinding. It would seem that any groove deeper than the marks which constitute the biting surface of the stone must necessarily afford some protection, because some fibers will gather in it and will be protected, so far as they are below the surface.
But we need not decide these disputed questions of fact. Hall’s invention, if he made one, was in supplying a new process and apparatus which produced better pulp, whether the reason for this result was because the grooves protected the fiber from regrinding or because they carried water, or because they had some other advantageous, but unexplained, effect. Both prior patents and prior use are relied upon to prove anticipation.
The closest prior patent is No. 1,118,671, to Millard, issued November ,24, 1914, for a method of dressing wood pulp grinding stones. His method consists in leveling the high spots on the face of a grindstone by rotating it against a smooth roller, “thereby developing the grit uniformly,” and then “in transversely and uniformly grooving and ribbing the leveled periphery of the stone from edge to edge to form a uniform series of transverse flat-topped ribs for grinding the wood and an alternating series of transverse grooves for escape of the ground pulp.” Here is a clear disclosure of the formation of grooves between the cutting ridges of the stone, just as in Hall’s process.
-The cutting surfaces between the grooves are treated differently. Millard uses a smooth roller to develop the natural grit of the stone. Hall says (page 2, line 25) that “the surface of the stone between the grooves is roughened in the usual manner.” He claims nothing with reference to such roughening, but says (page 1, lines 70-78) that it is old practice to sear with shallow depressions the surface between the grooves. Since the alleged novelty of Hall’s invention rests solely on grooving the stone, so as to afford protection to fibers, and since he claims no specific dimensions for his grooves, it is apparent that he might utilize the same grooves as Millard, who also specifies no dimensions. Hall (if his preferred dimensions be disregarded) adds nothing to the teaching of Millard with respect to the use of grooves, except to explain that grooving is advantageous, because it affords protection from regrinding. An explanation of why an old device works well is not invention.
The prior use relied upon is an even clearer anticipation. From August, 1911, to December, 1913, witnesses Thickens and MeNaughton carried on experiments in wood grinding at the Forest Products laboratory at Wausau, Wis. Their results were published by the Forest Service of the United States Department of Agriculture in pamphlet form *908on'June 11,1912, and July 12,1913, and have ever since been regularly on sale at the Government Printing Office. Defendant’s Exhibits 1 and 2. The method of dressing a stone used by them is clearly described and portrayed by plates in these pamphlets. It conforms very closely to the accepted practice which plaintiff claims as the teaching of Hall’s patent.’ On page 17 of Exhibit 1 appears the following:
“The stone was first dressed with a three to the inch burr, forming grooves in the stone approximately one thirty-seeond inch deep; then the portion of the stone between these depressions was roughed with a 12-cut spiral burr. This caused the grit to stand out and gave a maximum of useful grinding surface. The pulp obtained with this surface was almost entirely free from shives, and the fibers were long and fine. The surface of stone used during these tests is shown in Plate III, Figure 2.”
Page 11 of Exhibit 2 reads thus:
“The condition of the surface of the stone depends upon several factors. The size and sharpness of the individual particles of grit, the ease with which the binding material is worn away, and the manner of dressing the stone are important. In these tests but one stone was used, and variations in its surface were obtained by working it with steel rolls of different design. * * * The purpose of the depressions in the stone is primarily to provide a path by which the ground wood can leave it. It is possible that burrs of certain design will give a greater amount of grinding surface than others, and that the production will in this way be slightly increased.
“Plates III and IV show some of the bush rolls used in surfacing the stone for runs, the data of which are given in tables 3 and 4. The surface obtained by burring with the rolls, shown on plate III seemed to give more satisfactory results than any other thus far tried. The stone was first dressed with a three to the inch roll, and depressions were formed from one thirty-seeond to one-sixteenth of an inch deep. The stone was then rolled with a 12 to the inch spiral-cut burr,' until the spiral markings were plainly discernible. It is not at all essential that a spiral burr be used; any finely cut burr will give approximately the same results, the idea being, of course, to raise the grip of the stone. This is best done with a burr approaching the grit of the pulp stone in fineness. The important thing, so far as quality-is concerned, is to give the particles of grit the correct treatment, rather . than to form a deeply-grooved surface on the stone.”
• Plaintiff argues that Thickens nowhere •discloses in his pamphlets the idea of protection against regrinding by means of the grooves, and that the grooves shown, in his plates are too wide and shallow to give protection, despite his oral testimony that they do. They are no more shallow — in fact, less so — than the accepted practice which plaintiff claims to be within Hall’s patent. In width they appear to be scarcely an eighth of an inch, which was the preferred dimension stated by Hall. We finding nothing to show that the grooves would not be fiber-protecting to some extent. Doubtless they had sloping sides, and, as Williams says, the ideal groove for protection would have vertical sides, but this is a condition which can never be achieved in practice, and is not suggested by Hall.
• In our opinion, plaintiff’s process patent was shown to have been anticipated, both by the Millard patent and by Thickens’ operations between 1911 and 1913. Therefore the injunction against its infringement was erroneously granted.
We turn now to Hall’s burr patent, Ho. 1,219,299. The patentee states:
“My invention consists in a burr having transverse V-shaped teeth, separated by channels, the bottoms of which are substantially flat and serve not only to definitely limit the depth to which the teeth will be driven into the stone to produce the depressions, but also to secure alternate flat-topped ridges. • • *
“Such a burr may be used by an unskilled laborer without any danger of injury to the stone, and with the certainty of producing in the stone the desired surface, and always producing a definite uniform pattern. * * * ”
The single claim of the patent reads as follows:
“I claim as my invention: A burr for use in treating the surface of grindstones, having teeth upon its peripheral surface, said teeth being separated by channels, the bottoms of which are substantially flat and which serve to limit the depth to whieh said teeth may be driven into the surface of the grindstone and to secure alternate flat-top ridges.”
Burrs with inverted V-shaped teeth separated by V-shaped channels, such as Plaintiff’s Exhibit 23, were old in the art. The novelty of Hall’s burr lies in having its teeth separated by flat-bottomed channels, which serve to limit the depth of the cut and to secure alternate flat-top ridges in the stone. These two functions of the flat-bottomed channel are emphasized in those portions of the specifications already quoted. In the re*909quirement that the fiat-bottomed channel serve “to secure alternate flat-top ridges” one is reminded of the first element in the Millard patent, namely, the leveling of the stone by means of a flat roller. All the testimony, however, is to the effect that the bottom of the channel does not limit the depth of the cut or contact with the surface of the adjacent ridge. No one ever drives a burr to the hilt, and the testimony is overwhelming that, if it were so used, it would break down the stone and destroy the edges of the groove.
Walker’s testimony of the failure of plaintiff’s attempt to use its burr, designated as style No. 126A, in this manner, has already been mentioned. Potter and Crossett gave similar testimony. Even Williams says the tooth should not be driven in more than half its length, while all the other experts restrict the penetration to one-quarter or one-third of the length of the tooth. Consequently it is persuasively urged that the inclusion in the claim of the requirement that the flat-bottomed channel shall limit the depth of the cut and contact with the surface of the adjacent ridge renders the patent wholly inoperative. The plaintiff is faced with this dilemma: Either the patent is, as we think, invalid because inoperative, or, if it be deemed operative in the way claimed, the defendant has not infringed by selling burrs which are not to be driven to the hilt. Hence, the injunctive relief granted on the burr patent must also be reversed.
There remains the question whether that portion of the decree may stand which requires an accounting for damages due to the defendant’s acts while its license remained unrevoked. By stipulation defendant admitted the sale, without payment of royalty, to a nonlieensee of the process patent, of flat-bottom burrs, “said burrs being manufactured under and in accordance with the Hall patent, 1,219,299.” We have to decide whether a licensee may set up the invalidity of his licensor’s patent when sued for acts beyond the scope of his license, but performed during the license term. The doctrine of estoppel between licensee and patentee has not escaped criticism (Keener, Quasi Contracts, 37-39; Woodward, Quasi Contracts, § 62), but is too firmly established in the law to be questioned when the licensee is sued for acts within the scope of his license. Kinsman v. Parkhurst, 18 How. 289,15 L. Ed. 385; United States v. Harvey Steel Co., 196 U. S. 310, 25 S. Ct. 240, 49 L. Ed. 492. We see no reason to extend it, however, to acts outside the limits of the license. As Judge Denison said in Indiana Mfg. Co. v. Nichols & Shepard Co. (C. C.) 190 F. 579, 584:
“The estoppel must be mutual. The' licensee may not deny the patentee’s title to the monopoly; the patentee may not deny the licensee’s right to act under that monopoly. It is difficult to see how, when * * * both complainant and defendant agree that there is no contract in existence permitting the act in controversy, either party can be es-topped by a contract relating to something else. * * * It seems to me quite clear that the complaining patentee cannot, at the same time, maintain the position that the act of the defendant licensee, manufacturing what is said to be the patented article, is outside the conditions of the license, and, therefore, not authorized by the license, and also the position that his title to the monopoly is conceded by the license and, therefore, cannot he disputed. ~ * * ”
The authorities, as we understand them, sustain the defendant’s contention that the validity of the patent was open to question. See T. H. Symington Co. v. Nat. Malleable Castings Co., 257 F. 564 (D. C. Ill.); Tate v. B. & O. R. R. Co., 229 F. 141 (C. C. A. 4). As already pointed out, if the patent was not invalid, because inoperative, there was no infringement.
The defendant also admitted the sale, without payment of royalties, of round-bottom burrs to process licensees. If such burrs are an equivalent of the patented burr, as plaintiff contends, its remedy is a suit upon the license agreement (see St. Paul Plow Works v. Starling, 140 U. S. 184, 196, 11 S. Ct. 803, 35 L. Ed. 404), not a suit for infringement, since whatever the license covers is within the terms of the license agreement, and therefore the patentee is estopped to. deny that his licensee was privileged to make it.
Damages for infringement of the process patent are not recoverable from defendant as a contributory infringer, unless estoppel can be invoked to prevent an assertion of its invalidity. The defendant held no license under the process patent. As indicated by the authorities already cited, estoppel does not extend to acts outside the terms of the licensee’s contract. Whether the process licensees would be liable for royalties by reason of their use of round-bottom burrs purchased from defendant, and whether defendant’s incitement of them to refrain from paying process royalties may on any theory impose liability upon him, are questions not before us. On this record the plaintiff was entitled to no relief of any kind.
Accordingly, the decree is reversed, and the cause remanded, for entry of a decree in conformity with the views above expressed.