This appeal was instituted to a judgement of the District Court holding appellant, plaintiff below, not entitled to recover royalties claimed in the manufacture of phonographs under a contract executed December 7, 1925, between appellant's decedent, Bertram C. Kenyon, and the Automatic Musical Instrument Company, a Delaware corporation, predecessor of appellee.1 Pursuant to this contract a patent was applied for July 15, 1927, and issued September 27, 1932, Kenyon patent 1,879,693, under which Kenyon received from the Delaware corporation the royalty of $5.00 per phonograph, making a total of approximately $48,000 for machines manufactured, models P, Z, J, and K. The Delaware corporation went into receivership in 1931 and all of its assets, including the Kenyon patent, were sold in 1933 to the Automatic Musical Instrument Company, a newly organized Michigan corporation, predecessor of appellee. The record changer which was a main feature of the Kenyon patent and the only feature at issue here had proved unsatisfactory. After the organization of the Michigan corporation new models were designed and manufactured in 1935, the ARC constructed under Green patent, 2,104,032 and manufactured until May 31, 1946, and the 500 model, manufactured thereafter and covered by the Vanderzee patent application.

Shortly before his death October 16, 1941, Kenyon claimed royalties on machines manufactured under models ARC and 500, and this suit was instituted against the Michigan corporation January 22, 1945, seeking recovery for such royalties. In the court below appellee asserted that the contract of December 7, 1925, created a mere license, a right personal and nonassignable, and that the assignment of the patents by the receiver of the Delaware corporation imposed upon the Michigan corporation no obligation to pay royalties. This contention was sustained by the District Court, 63 F. Supp. 591, but reversed by this court, 160 F.2d 878 and the case proceeded to trail upon the merits.

It was conceded that appellee was liable for royalties in the period from 1935 to 1947 if its models ARC and 500 fell within the scope of the Kenyon patent. While the Kenyon patent covers several improvements in phonographic construction, the issue herein was narrowed by stipulation to the record-changer alone.

Evidence upon certain vital points is undisputed. It appears without contradiction that the Kenyon record-changer jammed in operation and broke many records, as a result of which models P, Z, J, and K were discarded and the tools, dies, and jigs for manufacturing them were scrapped. The new mechanisms claimed here to be covered by Kenyon were sharply differentiated from the old in structure and operation. As summarized by the District Court: 'In the Kenyon mechanism the record rack is stationary- in defendant's it is movable; in Kenyon the selector arm is swingable in a different path for each record- in defendant's the record-transfer arm moves in a single fixed path; in Kenyon the selector arm moves to the rack and selects the record- in defendant's the rack moves to a positive registration point on the fixed path of the arm, and the arm picks up whatever record is presented to it; in Kenyon the jaws pressure-grip opposite sides of the record at only one point- in defendant's, two brackets on the transfer arm engage the record at two points on opposite peripheral edges.'

It is stipulated that the new model ARC was manufactured in accordance with Green, 2,104,032, and that model 500 was manufactured under Vanderzee, but appellant claims that nevertheless the manufacture is in accordance with Kenyon and that appellee is therefore required to pay royalties to the time of the rejection of the contract in 1947.

Upon this, the principal point in the case, the District Court made the following findings of fact:

'5. The receiver of the Delaware corporation, acting in pursuance of court order, and the defendant company, designed and developed a new phonograph generally referred to as the ARC model, with a record-changing mechanism fundamentally and substantially different from that described in the Kenyon patent and from that employed by the Delaware corporation in models P, Z, J, and K prior to 1931; the defendant began the manufacture and sale of its new model ARC in 1934, and continued its manufacture until 1946, when it changed to model 500.

'6. The record-changing mechanisms in defendant's ARC and 500 models are fundamentally and substantially different from the record-changing mechanism described in the Kenyon patent and employed by the Delaware corporation in models P, Z, J, and K prior to 1931.

'7. The record-changing mechanisms employed by defendant in its ARC and 500 models are not within the scope and claims of the Kenyon patent or within the invention referred to in the 1925 contract.

'8. The record-changing mechanisms in defendant's ARC and 500 models are not equivalents of the mechanism described in the Kenyon patent and are not a mere reversal of the parts of the Kenyon mechanism.

'9. Although Bertram C. Kenyon knew of the development and manufacture of the ARC model, he made no claim for royalties until a few weeks before his death in 1941.'

The findings of the District Court are amply supported by the record, and we do not review the evidence upon these points, but proceed to consider the errors of law claimed to have been committed at the trial. For the detailed history of the case reference is made to the opinion of the District Court, 89 F. Supp. 602.

It is urged that the court erred in holding that the Kenyon invention was in a crowded field. However, record-changers were old in the art. Nelson, 1,595,241, and Seal 1,741,040, both disclose phonographic record-changers. Moreover, t he Kenyon patent included in its express objects 'to provide improved means for moving any desired one of a plurality of record disks to and from the turntable of the instrument; and further, to provide improved means whereby either of the records on the opposite sides of the disk may be selected to be 'played' * * * .' As repeatedly held, an inventor's appraisal of his own invention is important, and here it disposes of the argument that this was a pioneer invention. So far from being a device that performs a function never before performed (3 Walker on Patents, Deller's Ed., 1709), the Kenyon record-changer was said by its author to be a mere improvement.

While the assignee was estopped to deny the validity of the Kenyon patent, Stubnitz-Greene Spring Corp. v. Fort Pitt Bedding Co., 6 Cir., 110 F.2d 192, it was not error for the District Court to admit evidence of the prior art to illumine the scope of the Kenyon invention. Midland Steel Products Co. v. Clark Equipment Co., 6 Cir., 174 F.2d 541, 543.

It is urged that the patent notice estops appellee from raising the defense claimed here. The notice lists 14 patents under the following statement: 'This device is manufactured under one or more of the following patents under * * * .' Among the patents is Green, under which it is stipulated that model ARC was manufactured. This notice, required under Sec. 49, 35 U.S.C., 35 U.S.C.A. § 49 , is vague; but appellant urges that taken together with the failure of appellee to reassign the Kenyon patent and the payment of royalties under Kenyon, the notice conclusively establishes liability. Various cases are cited in which the existence of a patent notice, together with other circumstances, was held to have bearing upon the question of liability for royalties. Cf. Harley C. Loney Co. v. Perfect Equipment Corp., 7 Cir., 178 F.2d 165; Kant-Skore Piston Co. v. Sinclair Mfg. Corp., 6 Cir., 32 F.2d 882; Collis Co. v. Consolidated Machine Tool Corp., 8 Cir., 41 F.2d 641, 645.

We think these decisions are not controlling in the factual situation here presented. In each of the cited cases the identical, or substantially the same tool was manufactured during the period for which royalties were claimed. As stated in the Collis Co. case, 'this identical tool was for seventeen years manufactured' and royalties were paid upon the devices. Here, 1n 1933, the tools, dies, and jigs for manufacture under the Kenyon patent were scrapped, entirely new machine tools were set up and entirely new devices were manufactured. No royalties were paid thereafter. Moreover, Kenyon attended a meeting in Chicago in 1932 in connection with the reorganization of appellee, at which the engineer of the company described the improvements that had been made and the manner in which the property was conserved for the stockholders. It is a fair inference that this talk included a description of the new models. At any rate on February 10, 1934, Kenyon wrote to the appellee and expressed his interest in 'the new line of phonographs,' and in a possible declaration of dividends, but raised no claim for royalties until 1941. We think, as found by the District Court, the Kenyon contract did not cover the ARC and 500 models, and the inclusion of the patent number in a somewhat meaningless patent notice does not change the situation. Certainly the interpretation of the parties, including Kenyon, gives force to this conclusion.

No reversible error was committed in the admission of evidence.

The judgment of the District Court is affirmed.