Duplan Corp. v. Deering Milliken, Inc.

MEMORANDUM OF DECISION

This patent-antitrust litigation consisting of thirty-seven separate actions consolidated for purposes of trial has been tried to the court without a jury on the liability issues only, and in this memorandum of decision the court will record its findings of fact and conclusions of law in compliance with Rule 52(a), F.R.Civ.P. Jurisdiction is based on 28 U.S.C. §§ 1331, 1332, 1337, 1338, 2201 and 2202.

HISTORY OF THE LITIGATION

The first of the many complaints involved here was filed in the Spartanburg Division, United States District Court for the District of South Carolina, on August 8, 1968, as Case No. 68-705. In this original suit Peering Milliken Research Corporation (DMRC) and Moulinage et Retorderie de Chavanoz (Chavanoz) sought recovery of royalties alleged to be due by Textured Fibres, Inc., as a sub-licensee of DMRC which in turn was the exclusive use-licensee in the United States of certain apparatus and process patents issued to Chavanoz in the United States and relating to the false twist texturing of synthetic yams. 1 Similar suits were thereafter instituted by DMRC and Chavanoz against various other textile manufacturers (Throwsters) engaged in the yarn texturing business.

On November 25, 1969, the Duplan Corporation instituted in the United States District Court for the Southern District of New York the first of a series of actions by the Throwsters against DMRC and Chavanoz attacking the validity of the Chavanoz patents and asserting claims under the antitrust laws. Joined as defendants with DMRC and Chavanoz in these Throwster actions were Peering Milliken, Inc. (DMI), of which DMRC is a corporate subsidiary, Ateliers Roannais de Constructions Textiles (ARCT-France), a French manufacturer of textile machinery, and ARCT, Inc., a corporate subsidiary of ARCT-France organized under the laws of North Carolina for the purpose of selling in the United States the textile machinery manufactured by ARCTFrance and embodying the Chavanoz patents. These actions by the Throwsters were followed by countersuits and counterclaims by DMRC and Chavanoz for unpaid royalties and patent infringement.

At that time the thirty-seven actions were pending in the federal courts in South Carolina, North Carolina, Virginia and New York. After treatment by various United States District Courts and a Panel on MultiDistrict Litigation all of the cases, which by this time involved generally the same, basic issues of unpaid royalties, patent validity and infringement, patent misuse and alleged antitrust violations, were consolidated in the District of South Carolina in 1971 as Civil Action No. 71-306. The thirty-seven actions are listed in Appendix B attached hereto. 2

*658 Assigned originally to the Honorable Donald Russell, these cases were re-assigned to the Honorable Robert W. Hemp-hill of the District of South Carolina upon Judge Russell’s elevation to a seat on the Court of Appeals for the Fourth Circuit in 1971. Thereafter Judge Hemphill assumed charge of the litigation, held numerous hearings, ruled on innumerable motions involving procedural, evidentiary and summary judgment matters and personally presided over the taking of a massive volume of deposition testimony in this country and in France. A summary of the previously-reported rulings and decisions in the case is attached as Appendix C.

The prodigious work of Judge Hemphill is summarized in a footnote to one of his decisions, Duplan Corporation v. Deering Milliken, Inc., 400 F.Supp. 497 at page 502 (D.S.C.1975). That case was concerned with a recusal motion filed by counsel for DMRC, DMI and Chavanoz which after characteristically careful and painstaking consideration Judge Hemphill denied. The decision was not appealed. Because of the press of other duties (see 400 F.Supp. at page 526, Footnote 159) Judge Hemphill thereafter asked to be relieved from further duties in this litigation, and the same was assigned to this writer.

Pre-trial conferences were held at Raleigh, North Carolina, on March 26 and June 4, 1976. The Throwsters who were aligned in interest on one side of the case were designated as plaintiffs and the parties opposing the Throwsters, DMRC, DMI, Chavanoz, ARCT-France and ARCT, Inc., were designated as defendants and will be so referred to during the course of this memorandum. The trial which was commenced at Rock Hill, South Carolina, on June 14, 1976 consumed ninety-one trial days and with periodic recesses was concluded on February 11, 1977.

THE PARTIES AND THEIR ALIGNMENT

The plaintiffs are companies, or divisions of companies, whose businesses involve the processing of synthetic filament yarns in order to make them suitable for a wide variety of end uses. Each plaintiff is a corporation organized and existing under the laws of the state indicated below and conducts its principal texturing activities in the city indicated:

State of Principal Place

Plaintiff Incorporation of Business

Blanchard Yarn Company, Inc. Delaware Whitakers, N. C.

Burlington Industries, Inc. Delaware Greensboro, N. C.

Burkyarns, Inc. North Carolina Valdese, N. C.

Dixie Yarns, Inc. Tennessee Stanfield, N. C.

The Duplan Corporation Delaware Winston-Salem, N. C.

Frank lx & Sons Virginia Corporation New Jersey Charlottesville, Va.

Hemmerich Industries, Inc. Pennsylvania Denver, Pa.

Jonathan Logan, Inc. Delaware Spartanburg, S. C.

Lawrence Texturing Corporation (Division of Duplan) Lillington, N. C.

Leon-Ferenbaeh, Inc. Pennsylvania Johnson City, Tenn.

Madison Throwing Company (Division of Burlington) Madison, N. C.

National Spinning Company, Inc. New York Washington, N. C.

Olympia Industries, Inc. Delaware Tuscaloosa, Ala.

Reliable Silk Dyeing Company, Inc. New York New York, N. Y.

Schwarzenbach-Huber Company New Jersey Luray, Va.

Spring-Tex, Inc. North Carolina Gibsonville, N. C.

Texelastic Corporation North Carolina High Point, N. C.

Texfi Industries, Inc. Delaware Lumberton and New Bern, N. C.

United Merchants & Manufacturers, Inc. Delaware Cartersville, Ga. and South Carolina

*659 The defendants opposing the Throwster plaintiffs are the following parties:

1. DMRC is a South Carolina corporation and is a wholly-owned subsidiary of DMI. It is the successor to Deering Milliken Research Trust, and its newly-acquired corporate name is Milliken Research Corporation. Under a license agreement with Chavanoz, DMRC is the exclusive use licensee in the United States and Canada of the right to use the Chavanoz patents in suit and the right to grant sublicenses.

2. DMI is a Delaware corporation which maintains a place of business in New York, but its headquarters and much of its manufacturing operations are maintained in South Carolina. Its newly-acquired corporate name is Milliken & Company. DMI is a diversified textile manufacturer.

3. Chavanoz is a French “societe anonyme” with its principal place of business in Chavanoz, France. Chavanoz is the owner of the eight patents in suit.

4. ARCT-France is a French corporation with its principal place of business in Roanne, France. It is a manufacturer of various kinds of textile machinery, and under contractual arrangements with Chavanoz, ARCT-France is the exclusive licensee of the rights to make and sell the inventions made pursuant to the Chavanoz patents.

5. ARCT, Inc., is a North Carolina corporation with its principal place of business at Greensboro, North Carolina. It was formed in 1966 by ARCT-France for the purpose of distributing in the United States the textile machinery manufactured by ARCT-France.

In addition to the named defendants herein, other persons and concerns are alleged to have conspired with the defendants in one or more of the antitrust offenses charged, including: Leesona Corporation (formerly known as Universal Winding Company) (“Leesona”), the Permatwist Company (“Permatwist”), a partnership whose members are Warren A. Seem, Nicholas J. Stoddard, Fred Tecce and Harold P. Berger, and Whitin Machine Works (“Whitin”).

a. Leesona is a Massachusetts corporation with its principal place of business in Warwick, Rhode Island. Leesona is a manufacturer and seller of false twist and other textile machinery and has licensed patents and technology relating to false twist and post-treating.

b. Permatwist is a Pennsylvania partnership which has been engaged in the promotion and licensing of apparatus and processes relating to false twist and post-treating.

c. Whitin was a Massachusetts corporation engaged in the manufacture and sale of textile machinery with its principal place of business in Whitinsville, Massachusetts. Prior to 1966, Whitin purchased ARCT false twist machines from ARCT-France and was the exclusive distributor of such machines in the United States.

The acts alleged to have been done by the corporate defendants and their alleged co-conspirators have been carried out by their officers, directors or employees, including but not limited to, the following:

a. Norman C. Armitage was an officer of DMRC and from time to time an officer of DMI. Dr. Armitage, who was a lawyer, was in charge of the business, legal, policy and administrative aspects of DMRC's false twist licensing program from its inception until his death in 1972.

b. Leo M. Soep was a French “conseil en brevets” employed until about 1966 by Comptoir des Textiles Artificiéis, an affiliate of Chavanoz. Soep represented Chavanoz in connection with its false twist activities. After 1966.he became an independent conseil en brevets but continued to perform services for Chavanoz. From time to time Soep represented ARCT-France in connec *660 tion with its false twist activities. From 1966 until his death in 1971, Soep owned five per cent of the issued and outstanding shares of ARCT, Inc., and was a member of its board of directors.

c. Henri Crouzet was at all relevant times the president of ARCT-France, and also, from the time of its formation in 1966, the president and member of the board of directors of ARCT, Inc.

d. Yves de Moncuit was at all relevant times an officer of Chavanoz and the Chavanoz employee who worked with Leo Soep on false twist matters. In 1969 or 1970 he became president of Chavanoz.

e. Robert F. Waters handled all ARCT false twist machinery sales for Whitin from 1959, when the first ARCT false twist machines were introduced in the United States, through 1965. Since the formation of ARCT, Inc. in 1966, Waters has been its executive vice president in actual control of its day-to-day operations, a member of its board of directors, a shareholder, and has continued to be the prime salesman for ARCT false twist machinery in the United States.

f. Walter E. Mueller was chief house patent counsel for DMRC (and its predecessor Deering Milliken Research Trust) from April, 1951 to March 1, 1968.

g. Robert Leeson, at all relevant times until 1967, was the president and chief executive officer of Leesona. Mr. Leeson was also chairman of the board of Leesona from 1956 or 1957 until at least February, 1972.

h. Albert P. Davis was at all relevant times the house patent counsel for Leesona.

THE PLEADINGS

At one time there were twenty-two Chavanoz patents in suit. Prior to trial twelve of the patents had been held not infringed on motions for summary judgment, and two of the patents had been held invalid under 32 U.S.C. § 102(d). Eight patents remain in suit. They are United States Patents Nos.

2,891,375 3,165,881

2,944,319 3,232,037

3,012,397 3,283,414

3,123,973 3,584,450

Plaintiffs allege in complaints, counterclaims and affirmative defenses against DMRC and Chavanoz that these remaining Chavanoz patents are invalid and not infringed. They seek declaratory judgments of invalidity and non-infringement. They deny any liability for royalties or for patent infringement. They also allege that DMRC, DMI, Chavanoz, ARCT-France and ARCT, Inc., have violated Sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1 and 2, and have committed acts of patent misuse.

The antitrust claims fall into two categories: (a) the claim of a vertical conspiracy between the named defendants arising out of the license and sub-license agreements and the manner in which the business of the defendants was conducted, and (b) a claim of horizontal conspiracy between the named defendants and Leesona arising out of agreements entered into in 1964 which settled certain patent litigation then pending between Leesona and the defendants. The misuse claims relate to proceedings before the Patent Office, license provisions and the conduct of the present litigation by the defendants.

As previously indicated, the first of the actions here involved was brought by DMRC against Textured Fibres, Inc. (now Texfi Industries, Inc.), on August 8,1968, as a simple contract action for the recovery of alleged unpaid royalties in the amount of $45,691.45. Following the institution of the suit by Duplan Corporation on November 25,1969, seeking a declaratory judgment of invalidity and non-infringement as to the Chavanoz patents and alleging antitrust violations and patent misuse DMRC asserted claims or counterclaims against each of the plaintiffs for breach of the sub-license agreements arising out of plaintiffs’ refusal to pay royalties and infringement of the Chavanoz patents arising out of the plaintiffs’ continued use of the ARCT machines embodying the Chavanoz patents following repudiation or termination by the plaintiffs of the sub-license agreements. Chavanoz, which is not a party to the sub-license agreements, has joined DMRC in asserting *661 claims of patent infringement and, of course, is defending against plaintiffs’ claims of antitrust violations and patent misuse.

ARCT-France, ARCT, Inc., and DMI disclaim any interest in or to the patents in issue and therefore make no claims for infringement damages or royalties against the plaintiffs. In their answers these three defendants have denied any liability to the plaintiffs by reason of the alleged antitrust violations.

THE ISSUES

All questions of damages, if any, to which any party may be found entitled having been reserved for trial at a later time, the issues arising on the pleadings to be resolved by the court at this time fall into four categories:

I. ANTITRUST

II. PATENT MISUSE

III. PATENT VALIDITY AND INFRINGEMENT

IV. NONPAYMENT OF ROYALTIES

These issues will be treated in the succeeding sections of the memorandum.

I.

THE ANTITRUST ISSUES

A. The Vertical Conspiracy.

The facts on which the plaintiffs’ allegations of antitrust conspiracy between the parties defendant in the chain of distribution are based are not in substantial dispute, the plaintiffs contending that the alleged restraints of a vertical nature are set forth in writing in the instruments the defendants signed. A review of the essential provisions of these agreements is therefore in order.

1. The Chavanoz-ARCT Agreements.

On October 30, 1954, Chavanoz entered into an agreement with ARCT-France (The “1954 Agreement”) under the terms of which Chavanoz granted to ARCT-France “the exclusive right to manufacture and sell” the inventions described in certain Chavanoz false twist patents and patent applications as well as any later improvements. At that time Chavanoz owned French Patent No. 1,054,338 and two applications (Nos. 52,346 and 54,253) for certificates of addition, but it owned no United States patents relating to false twist.

The pertinent provisions of the 1954 agreement were as follows:

“1. [Chavanoz] grants to ARCT the exclusive right to manufacture and sell the devices described in the patent and additions mentioned above as well as any later improvement.

“4. . . . [T]he rights of industrial property are reserved exclusively for [Chavanoz] including those attached to the new models. Any application for a patent that could concern such new models must be made by [Chavanoz] in its name and at its expense. ARCT will take the necessary steps to inform [Chavanoz] promptly of any improvements so as to allow it to insure adequate protection of which [Chavanoz] is the sole judge.

“6. ARCT shall not deliver the patented material to any firms other than licensees of the HELANCA process, except in the case of a prior and written authorization from [Chavanoz]. 3

“7. In exchange for the present exclusive grant, ARCT shall pay to [Chavanoz] royalties on all the material built and invoiced by ARCT or by its sub-licensees *662 by virtue of the present document, delivered to any firms other than [Chavanoz], The royalties shall be as follows . [ten per cent for first year decreasing annually to two per cent for the sixth and following years].

“9. The present license for construction and sale is granted and accepted for the duration of the main patent [French Patent No. 1,054,338], and it shall come into effect on the day the first mass-produced machine following the prototype is delivered.”

Thereafter Chavanoz obtained United States patents corresponding to the French patent and applications which were issued as United States Patents Nos. 2,741,893 (the “bathtub” patent — so-called because of the similarity in shape of the vessel designed to contain a hot liquid through which the yarn passed during processing to a conventional bathtub); 2,761,272; 2,780,047; 2,823,513; and 2,823,514. Since the 1954 agreement covered “improvements” as well as the existing French patent and applications and was worldwide in scope, these and subsequent United States patents obtained by Chavanoz relating to false twist came under the 1954 agreement. All five of these patents were held by Judge Hemphill on motions for summary judgment to be non-infringed by any ARCT machine purchased by the plaintiffs in this action.

Chavanoz and ARCT-France entered into a revised agreement dated July 18, 1962 (The “1962 Agreement”) which consolidated “in a comprehensive instrument” the 1954 agreement and a “number of verbal agreements” stating “the parties are at one in considering that the various agreements should be brought into line and consolidated in a comprehensive instrument.” The essential provisions of the 1962 agreement were as follows:

“1. [Chavanoz] grants to [ARCT] the exclusive right of construction and sale in the whole world of the devices described in the patents and patents of addition listed on the attached List A, as well as of any subsequent improvement in the field of the manufacture of crimped textile yarns, either natural, artificial or synthetic, curled through the application of a false twist which is heat set, and in the yarns which may be obtained from said crimped or curled yarns, through an additional treatment or shaping. This field is called the ‘field of the patents’.

“4. All patent rights attached to the improvements in the field of the patents remain the property of [Chavanoz]. Any patent application which may concern new models must be filed by [Chavanoz] in his name and at his expense, and [ARCT] must take all necessary measures so as to diligently communicate the improvements to [Chavanoz] to enable him to assure adequate protection, of which [Chavanoz] remains the only judge.

“7. [ARCT] has filed in its name the patents listed on List B and the patents have been assigned to [Chavanoz] with a retrocession of the complete freedom of exploitation outside of the field of the patents . . .

“8. [ARCT] shall deliver the devices under the present contract only to the ■holders of a process license which is granted by [Chavanoz], except in the case of a prior and written authorization by [Chavanoz], In the countries where [Chavanoz] has no patents, [ARCT] may deliver without authorization.

“9. [ARCT] pays to [Chavanoz] in exchange for the exclusive right of construction and sale according to Article 1 and the technical assistance according to Article 2 [‘the drawings and models of the prototypes made by the patentee’], a royalty which is two per cent of the value before taxes, leaving factory, electric motors not included.

“11. The present license of construction and sale is granted and accepted for the duration of the French Patent No. 1,054,338 [U.S. Patent No. 2,741,893 — the “bathtub” patent].

“19. By reason of the technical assistance supplied by [Chavanoz] according to Articles 1 and 2, the present contract is not cancelled in the case of the complete invalidity of the patents of List A, but in *663 this case the protection according to Article 15 is de facto eliminated.”

An additional French patent of addition (No. 67151) issued subsequent to the 1954 agreement was listed in List A attached to the 1962 agreement, but apparently no United States patent was issued corresponding to it. Of the six French patents assigned by ARCT to Chavanoz as shown on List B, United States patents were obtained on two of them (French Patent Nos. 1,126,-065 and 1,216,847) and these two United States Patents, Nos. 2,788,634 and 3,177,361, were held by Judge Hemphill not to have been infringed by the plaintiffs herein.

The royalties payable by ARCT to Chavanoz under the 1954 and 1962 agreements were in fact paid, and there was no difference in the royalty rate for machines sold in the United States where customers also paid a production royalty for use of the machines and the royalty rate paid on machines sold in countries where no use royalty was payable.

2. The Chavanoz-DMRC Agreements.

On December 31,1957, DMRC and Chavanoz signed an agreement giving DMRC the exclusive use rights in the United States and Canada under the Chavanoz false twist patents with the right to grant sub-licenses. The pertinent portions of this agreement were as follows:

“1. Chavanoz hereby grants to DMRC throughout the United States, its territories and dependencies and Canada upon the conditions hereinafter set forth and subject to certain rights of cancellation as defined below, an exclusive license to use the FT process and FT machines for the purpose of making, using and selling [false twist] yarns in accordance with technical information and the inventions of the FT patents set forth in Appendix ‘A’ appended hereto, with the right to grant sublicenses in the United States and Canada. The right to manufacture and sell FT machines in accordance with said FT patents is specifically excepted.

“3. The manufacture and delivery of FT machines by ARCT or its sublicensees for use in the United States and Canada shall, as between Chavanoz and DMRC, be the responsibility of Chavanoz and Chavanoz shall direct ARCT or ARCT’s sublicensees for the manufacture and sale of FT machines to sell or deliver FT machines in the United States and Canada only to parties sublicensed by DMRC to use the machines, such sublicenses and parties being respectively referred to hereinafter as ‘use licenses’ and ‘use licensees’ .

“DMRC under its rights to sublicense hereunder shall issue use licenses to reputable customers of ARCT or of ARCT’s sublicensees when called upon by ARCT or ARCT’s sublicensees so to do, but shall have the right to refuse the grant of a use license to any parties for sufficient cause. The grant of a use license shall not be arbitrarily or unreasonably withheld . . .

“9. DMRC shall charge its use licensees royalties in the amount of five per cent of the manufacturer’s list price of the raw yarn which is converted to mousse [false twist] yarns by such use licensees according to the FT process or on FT machines and is sold or used; provided, however, that this rate of royalty may be reduced by DMRC and at its discretion to a figure of not less than two and one-half per cent if there is substantial unlicensed competition by producers using the FT process or FT machines embodying the inventions of this agreement to the extent of at least 10,000 pounds of [false twist] yarn per month and further provided, that DMRC shall exact from each use licensee a minimum annual royalty of $1,000.00 payable in advance.

“10. DMRC shall remit to Chavanoz fifty per cent of all revenues which DMRC itself receives from the use licensees .

“15. This agreement and the license granted hereunder, unless sooner terminated or cancelled as hereinafter provided, shall continue for a period until expiration of the last patent to issue to Cha *664 vanoz in the United States and Canada respectively.”

In addition to the quoted provisions the agreement also required Chavanoz “to furnish to DMRC all the technical information and know-how Chavanoz possesses in the field of the FT process”, and there was a grant-back provision requiring DMRC to assign to Chavanoz improvements in the FT process made by DMRC and to include in its sublicense agreements a provision requiring its sub-licensees to grant Chavanoz a license in the sub-licensees’ own country under any such improvements and to assign all foreign rights to Chavanoz without payment of royalty.

DMRC and Chavanoz executed six supplemental agreements between 1957 and 1962 under the terms of which Mexico was added to the territory in which DMRC was given use rights, the grant-back provision in the 1957 agreement was first modified and then deleted, the royalty rate was fixed at two and one-half per cent and DMRC was authorized to grant non-commercial use licenses with no minimum royalty. A further supplemental agreement provided that the sub-licenses granted by DMRC would not ipso facto terminate “if for any reason the exclusive license granted DMRC by Chavanoz is terminated.”

On December 28,1962, DMRC and Chavanoz entered into a new basic agreement which revised the 1957 agreement, but in all material respects it remained the same as the 1957 agreement as modified by the intervening supplemental agreements.

With a few subsequent modifications, the 1962 DMRC-Chavanoz agreement remained the basic agreement between them until after the commencement of the present litigation. One of the modifications incorporated the terms of an agreement between DMRC, Chavanoz and Whitin in June, 1963 increasing the royalty rate to three and one-half per cent with all but .6% being escrowed for return to the licensees in the event of the unsuccessful defense of litigation then pending between the Leesona Corporation, DMRC, Chavanoz and Whitin.

3. The ARCT-France-Whitin Agreement.

On February 20,1959, ARCT-France and Whitin Machine Works entered into an agreement under which Whitin was granted the exclusive right to sell ARCT false twist machines in the United States, Canada and Mexico. (An option granted Whitin to manufacture the machines was never exercised.) Following several recitals including reference to the 1954 agreement between Chavanoz and ARCT-France and the fact that Chavanoz has granted to DMRC “the exclusive right and the right to grant sub-licenses to practice and use the processes covered by said [Chavanoz] patents in the USA, Canada and Mexico” the agreement contained the provision that “Whitin agrees to sell FT machines only to persons or firms approved by [Chavanoz] and/or DMRC.”

While the agreement speaks in terms of Whitin’s acting “as a selling agent for such machines”, in practice Whitin purchased the machines outright from ARCT-France and took title to them at the French port of embarkation. At that time ARCT-France parted with all dominion and control over the machines and the risk of loss was transferred to Whitin. Whitin had the absolute right to set its own price for the resale of these machines to Throwsters in the United States, Canada and Mexico.

4. The ARCT-France — ARCT, Inc., Agreement.

In February, 1966, ARCT, Inc., was organized as a North Carolina corporation to undertake to sell the ARCT machines in the United States. Stock in this new corporation was owned sixty per cent by ARCTFrance, thirty-five per cent by Robert Waters, who had been Whitin’s sales manager for the ARCT machines, and five per cent by Leo Soep, a French “conseil en brevets” 4 , who, in addition to representing *665 Chavanoz in patent matters, also negotiated agreements in patent and other matters on behalf of ARCT-France from time to time including the Whitin agreement.

With the formation of ARCT, Inc., Robert Waters left Whitin, where he had been responsible for the sale of all ARCT false twist machines in this country, to become executive vice president and director of ARCT, Inc., in actual control of its day-today operations.

On February 7, 1966, ARCT-France and ARCT, Inc., entered into an agreement for the purchase of ARCT machines from ARCT-France by ARCT, Inc. This agreement provided that “property in the machine and risk of loss will shift to ARCT, Inc., upon delivery to the ocean carrier at the French port of embarkation” and that ARCT, Inc., had the absolute right to set its own resale price to its Throwster customers. Although Whitin retained the right to distribute the machines after the formation of ARCT, Inc., it in fact went out of that business and did not sell any false twist machines thereafter, and since 1966 ARCT, Inc., has acted as the distributor of ARCT false twist machines in the United States.

The written agreement between ARCTFrance and ARCT, Inc., contained no express covenant, such as that found in the ARCT-Franee-Whitin agreement, restricting the resale of ARCT machines by ARCT, Inc., to DMRC licensees, but in practice ARCT, Inc., did in fact so restrict delivery of the machines until well after the institution of this litigation.

The sales contracts of ARCT, Inc., and its predecessor, Whitin, contained no reference to the DMRC use license or the fact that DMRC had any use rights in the Chavanoz patents, but the Throwster purchasers were routinely informed by Waters while he was employed by Whitin and later by ARCT, Inc., that it would be necessary to obtain a use license from DMRC before the machinery could be placed in operation. It was also publicly announced in trade publications as early as April, 1959, that Chavanoz had granted DMRC the right to license users of the ARCT machines in the United States which Whitin had been licensed to manufacture and sell, and it was common knowledge in the trade that a DMRC use license was required to operate these machines.

5. The Standard DMRC License Agreement.

Prior to the sale of the first ARCT machine in the United States DMRC prepared with Chavanoz’s approval a printed form standard license agreement (the “DMRC Use License”) to be signed by all purchasers of ARCT machines. The subject matter of the DMRC use license, which was signed by each of the plaintiffs in substantially identical form, is spelled out in the “Whereas” clauses as follows:

“WHEREAS, DMRC has an exclusive license throughout the United States, Canada and Mexico with the right to a grant sublicenses under certain inventions and technical information relating to processes and devices for the manufacture of crimped synthetic yarns based upon the application of a false twist (such processes and devices being hereinafter referred to respectively as ‘FT processes’ and ‘FT machines’), which inventions are described in United States patents and/or applications for Letters Patent in the United States, owned by MOULINAGE ET RETORDERIE DE CHAVANOZ (hereinafter referred to as CHAVANOZ) and listed in Appendix ‘A’ appended hereto, together with certain improvements thereon as such may hereafter be made or acquired by CHAVANOZ and any patent applications and/or patents in the United States relating thereto, to use said processes and devices for the purpose of making said crimped yarns for use and sale, such right of DMRC under the inventions, applications and patents aforesaid being hereinafter referred to as ‘FT PATENT RIGHTS,’ and

*666 “WHEREAS, LICENSEE desires a use license to use processes and devices embodying the inventions of said FT PATENT RIGHTS, ...”

The pertinent contractual provisions of the agreement are as follows:

“1. DMRC hereby grants to LICENSEE upon the conditions hereinafter set forth and subject to certain rights of cancellation as defined below, a nonexclusive and nontransferable use license for a period until expiration of the last patent to issue in the United States, upon which said FT PATENT RIGHTS are based, to use the FT processes and FT machines for the purpose of making for use and sale crimped yarns in accordance with technical information and the inventions of said FT PATENT RIGHTS. The use license hereby granted is restricted as to the use of the FT process and FT machines to the plants of the LICENSEE situated in the United States, but subject to intervening rights, if any, of third parties, the crimped yarn manufactured by such use may, as between DMRC and LICENSEE, be sold freely in all of the countries of the world.

“2. DMRC has already furnished to LICENSEE certain technical information relative to the present inventions, which LICENSEE acknowledges, and as promptly as practicable after the date of this agreement DMRC shall furnish to LICENSEE such additional technical information and ‘know-how’ as is necessary in DMRC’s opinion to enable LICENSEE to practice the inventions licensed hereby and shall from time to time while this agreement is in effect furnish further additional information as it similarly deems necessary to supplement information heretofore furnished hereunder, provided, however, there shall be no obligation on the part of DMRC or its licensor to perform any additional or future research or development in the field of the inventions covered by this agreement. DMRC shall disclose to LICENSEE said additional information and ‘know-how’ after mill test has in DMRC’s judgment confirmed that an improvement has been made and within sixty (60) days after the improvement has in DMRC’s judgment been successfully reduced to practice in commercial production. If DMRC hereafter makes or acquires any improvements of the inventions of FT PATENT RIGHTS upon which it obtains patents, it shall then grant to LICENSEE licenses to use such improvements at no increase in royalty by incorporation of such patents into FT PATENT RIGHTS under the present agreement. LICENSEE shall be entitled to send its engineers or other personnel to DMRC or its designee for the purpose of obtaining instructions as to the best methods of practicing these inventions, and may request DMRC to send to LICENSEE upon terms to be mutually agreed upon technical personnel for the purpose of instructing LICENSEE at LICENSEE’S premises in the said best methods of practicing the inventions.

“3. LICENSEE shall disclose to DMRC within thirty (30) days of the first use or embodiment thereof in commercial practice, any improvements of the FT process or FT machines, whether or not patentable, conceived and made by LICENSEE or its employees subsequent to the date of this agreement and shall grant to DMRC or its designee throughout the United States, its territories and dependencies, a nonexclusive license thereunder and any patent application or patent thereon with the exclusive right in DMRC or its designee to sublicense its licensees and sublicensees thereunder for the life of the last patent to issue in the United States of the patents upon which said FT PATENT RIGHTS are based, said improvements being available for use by LICENSEE, and LICENSEE shall assign to DMRC or its designee all foreign rights thereto, all without payment of royalties. [This grantback clause deleted after 1961.]

“4. Except as hereinafter provided, LICENSEE shall pay DMRC during the life of this agreement royalties in the amount of two and one-half per cent (2%%) [later 3%%] of the manufacturer’s list *667 price of the raw yarn (but including any customs tariff on yarn imported from abroad) which is converted to crimped yarn by LICENSEE according to the FT process or on FT machines and is sold . The present use license is related only to the use of FT machines manufactured under license of CHAVANOZ by Ateliers Roannais de Constructions Textiles, of Roanne, France, referred to hereinafter as ‘ARCT,’ or its sublicensees, and LICENSEE is required to pay royalties under the provisions of the present paragraph only upon the production of such machines, provided that LICENSEE shall have the right to include under this use license the use of any other false twist or FT process or any other false twist or FT machine upon notice to DMRC, whereupon royalties upon the production thereof shall thereafter be payable in accordance with the above provisions.

“5. LICENSEE shall pay to DMRC a minimum annual royalty of One Thousand Dollars ($1,000.00) in United States currency, the first said payment to be made upon the signing of this agreement by LICENSEE and succeeding payments upon each anniversary of the execution of the agreement. These minimum royalties in their entirety shall be respectively credited against royalties accruing under paragraph 4 hereof in the next succeeding twelve (12) months but shall not be credited against any royalties payable thereafter.

“9. DMRC warrants that CHAVA-NOZ has undertaken that, in the event that LICENSEE is threatened with suit or is sued for patent infringement based upon the use of techniques or procedures specifically recommended by DMRC to LICENSEE hereunder, CHAVANOZ shall upon request from LICENSEE (transmitted through DMRC) defend such suit at the expense of CHAVANOZ insofar as such alleged patent infringing activities may be involved; provided that DMRC is notified promptly in writing of all such claims of or suits for infringement, and further provided that any damages awarded or expenses of any kind incurred in such defense beyond court costs and attorneys’ fees shall be borne by LICENSEE.

“10. DMRC further warrants that CHAVANOZ has agreed that, if for any reason the exclusive license granted DMRC by CHAVANOZ and referred to above is terminated, the present use license shall not be terminated ipso facto but LICENSEE shall have the right and option to continue under the terms of the present use license, except that CHAVA-NOZ shall succeed to the rights and obligations of DMRC hereunder.

“11. LICENSEE shall have the right to terminate this license and agreement five (5) years after the date of commencement of the first fiscal year hereunder or on any anniversary thereafter by giving DMRC sixty (60) days’ notice in writing of such termination .

“13. In the event of termination or cancellation of this agreement by operation of paragraphs 10, 11 or 12 hereof, LICENSEE agrees to cease using the FT process and FT machines for the manufacture of crimped yarn according to inventions which are the subject of the present agreement, except that LICENSEE shall have the right to complete any and all contracts for the manufacture of said crimped yarn which it may then have upon its books or for which it has become obligated ... In the event of termination or cancellation of this agreement, LICENSEE shall not use or disclose the technical information furnished hereunder except as such information is published or otherwise made available to the public through other sources, and LICENSEE shall deliver to DMRC within ninety (90) days after the date of such termination or cancellation all written or printed material in LICENSEE’S possession relating to the FT process or FT machines of this agreement, whether or not such written or printed material was furnished to LICENSEE by DMRC, and including all copies of instructions, drawings, photographs, and the like.

*668 “14. LICENSEE acknowledges the validity of any patents issued or which may issue on applications as aforesaid, and agrees that it will not contest the same or be a party directly or indirectly to any proceeding disputing such validity or tending to impair the value of FT PATENT RIGHTS or by which the enjoyment of full revenue therefrom by DMRC may be reduced.

From the outset of the DMRC licensing program in the United States DMRC and Whitin actively cooperated in compelling compliance with the use license requirement, and this cooperation was continued by ARCT, Inc., when it took over the sales of the ARCT machines in this country. DMRC took such measures as threatening to embargo the shipment of ARCT machines to the United States in an effort to prevent the delivery of machines to non-licensed Throwsters.

The procedures which had been established during the Whitin period continued essentially unchanged by ARCT, Inc., until after the commencement of this litigation. Although there were some isolated instances in which machines were delivered prior to the execution by the purchaser of the use license, Waters never told a customer or prospective customer that it was not necessary to sign the DMRC license or that he would deliver a machine if the agreement was not signed.

ARCT-France through its chief executive officer, Henri Crouzet, also continued to cooperate with Chavanoz and DMRC in the use licensing program. As late as February, 1970, Crouzet wired Armitage of DMRC:

“. . .It has never been in our intention nor Bob Waters’ to deliver machines to your customers without signature of a license . . . The agreement given to Bob Waters is to sell at the present conditions up to end of February to the new customers who would have taken towards ARCT the binding of signing a license with you . . . The license will always been [sic] regularized before delivery of the machines . . . ” (PX 539).

And in March of 1970 Crouzet wrote to deMoncuit of Chavanoz:

“Pursuant to Article 8 of the agreement of 7/18/62 we refrained from and prohibited our affiliate, ARCT, Inc., from selling false twist machines to American customers who have not taken a license with DMRC.” (PX 212).

The result of this concert of action between the defendants was that at the time of the institution of this litigation there was no ARCT FT machine in commercial operation in the United States by an unlicensed user.

Because the requirement that the machines be sold only to use licensees was a disadvantage to Whitin and ARCT, Inc. in making sales, the cooperation of Robert Waters in the DMRC licensing program while he was sales manager for Whitin and later when he became executive vice president of ARCT, Inc. was tinged with some reluctance. As shown in more detail in the proposed findings of fact of ARCT, Inc., Nos. 51-52 adopted below, this eventually led to a deterioration of the relationship between Waters, whose sole interest was in selling machines, and Norman Armitage of DMRC, whose sole interest was in licensing the machines and collecting royalties. Finally in late August, 1970, long after this litigation had been pending, ARCT, Inc., added a disclaimer paragraph to its sales contracts reading as follows:

“There is no warranty, express or implied, that the sale, delivery- or use of the FT machine provided for in this contract does not infringe patents owned by third parties. A use license under certain patents covering this machine may be obtained from Deering Milliken Research Corporation.” (DX 821).

Thereafter ARCT, Inc., sold the machines without regard to whether the customer had signed a use license, and no new purchaser has since signed a DMRC use license.

In addition to the foregoing findings of fact with respect to the alleged vertical conspiracy the court expressly adopts as its *669 own the following proposed findings of fact submitted by the parties:

1. Plaintiffs’ proposed findings of fact on the antitrust issues Nos. 14.40, 14.43, 14.46, 14.51, 14.56, 14.59, 14.60, 14.61, 14.64, 14.66, 14.80, 14.81 and 14.82.

2. Chavanoz, DMRC and DMI proposed findings of fact on the antitrust issues under Section I, Nos. 1-9 inclusive, 11, 12, 20-23 inclusive, 29, 33, 72 and 73.

3. ARCT-France’s proposed findings of fact Nos. 7-12 inclusive, 16, 17, 24, 25, 27 and 30-33 inclusive.

4. ARCT, Inc.’s proposed findings of fact Nos. 50, 51 (with the exception of the last paragraph) and 52.

While the foregoing facts relating to the alleged vertical conspiracy are not in serious dispute, the legal conclusions drawn by the opposing parties from these facts are in diametric contradiction. The plaintiffs have confidently asserted that they have established by a preponderance of the evidence per se and other violations of Section 1 of the Sherman Act while the defendants with equal confidence have contended to the contrary, asserting that the facts establish only that the defendants have exercised lawful rights granted them under the patent laws. 5

Plaintiffs’ position may be briefly summarized as follows:

(1) Chavanoz’s license to ARCT-France and its sale of the machines exhausted the patent monopoly and gave all subsequent purchasers of the machines an implied license to use them without further payment of royalties;

(2) The Chavanoz-ARCT-France agreements required ARCT-France to assign to Chavanoz (“grant-back”) patent rights in all improvements;

(3) Chavanoz and DMRC conspired to fix the price of the ARCT machines through the use royalties exacted from the Throwster purchasers; and

(4) Chavanoz and DMRC conspired to tie the purchase of the machines to the purchase of a compulsory package license covering unpatented “technical information” and technology and a large number of patents most of which were not applicable to the machines.

These charges will now be considered seriatim.

The Exhaustion-Implied License Theory

Plaintiffs do not deny that the rights inuring to a patentee under Section 154 of the patent laws, 35 U.S.C. § 154, to exclude others from making, using or selling a patented invention may lawfully be assigned or licensed separately under Section 261, 35 U.S.C. § 261, and that the patentee is entitled to a monetary reward for any one or all three of such rights. Plaintiffs earnestly contend, however, that Chavanoz failed to achieve its apparent purpose in this case with the result that the sale of the ARCT machines, first to Whitin and later to ARCT, Inc., exhausted the patent monopoly and that upon resale of the machines the purchasers acquired an implied right to use them without payment of a use royalty to Chavanoz or its sublicensee, DMRC. From this premise it is argued that the restriction on the resale of the machines to DMRC licensees was a restraint on trade which constituted a per se violation of Section 1 of the Sherman Act under United States v. Arnold, Schwinn and Company, 388 U.S. 365, 87 S.Ct. 1856, 18 L.Ed.2d 1249 (1967).

In support of their argument that the exhaustion doctrine is applicable here plain *670 tiffs point to the absence of any express reservation of use rights in the 1954 agreement. This agreement gave ARCT-France, so the argument goes, the unlimited right to sell the machines free and clear of any use rights later claimed by Chavanoz to have been retained, and notwithstanding all purchasers of the machines in the United States fully understood that use royalties were payable, and were in fact paid prior to the institution of this litigation, the purchasers were under no legal obligation to do so. Settled principles of contract and patent law impel a contrary conclusion.

The general rules of construction for contracts are applicable to the construction of patent licenses. De Stubner v. United Carbon Company, 67 F.Supp. 884, 891 (S.D.W.Va.1946), aff’d, 163 F.2d 735 (4th Cir. 1947); Baldwin Rubber Company v. Paine & Williams Company, 107 F.2d 350 (6th Cir. 1939). The construction placed on a license contract by the parties is entitled to great weight. Limbershaft Sales Corporation v. A. G. Spalding & Brothers, 111 F.2d 675 (2nd Cir. 1940). And, of course, a license contract must be construed as a whole and the intention of the parties must be determined from the entire agreement. Victory Bottle Capping Machine Company v. O. & J. Machinery Company, 280 F. 753, 759 (1st Cir. 1922).

None of the agreements between Chavanoz and ARCT-France expressly reserved to Chavanoz the use rights in machines to be built embodying the Chavanoz inventions, and sales were only limited by the 1954 agreement to licensees of the HELANCA process. Nothing was said of the right of those licensees to use the machines following purchase. 6 The 1957 Chavanoz-DMRC agreement, however, obligated Chavanoz “to direct ARCT ... to sell or deliver FT machines in the United States only to parties sublicensed by DMRC to use the machines . . .”

The record is not clear as to how Chavanoz undertook to discharge this obligation, but without question ARCT-France and its affiliates in the United States, first Witin and later ARCT, Inc., faithfully adhered to the directive which Chavanoz doubtless gave, and no sales or deliveries were made to purchasers in the United States who did not at the time of the sale or shortly thereafter sign a DMRC use license. It must be assumed, therefore, that ARCT-France accepted this obligation as an amendment to its 1954 agreement with Chavanoz. The 1962 agreement which consolidated “in a comprehensive instrument” the 1954 agreement and a “number of verbal agreements” contained an express prohibition against the delivery of the machines by ARCT-France to non-licensed users. In the meantime a similar prohibition had been incorporated in the 1959 agreement between ARCT-France and Whitin.

Even if there had been no express limitation on ARCT-France’s right to sell and deliver to non-licensed users, such term may be implied from the conduct of the parties.

“Terms may be implied in a contract, not because they are reasonable, but because they are necessarily involved in the contractual relationship so that the parties must have intended but failed to specifically include them because of their obviousness. Where, from the nature of a contract and the circumstances under which made, it is apparent the parties must have proceeded on the basis that certain conditions existed, without which its performance would be unnecessary, the existence of such conditions will be regarded as implied terms of the obligation. Sacramento Navigation Company v. Sate, 273 U.S. 326, 329 [47 S.Ct. 368, 71 L.Ed. 663]; Wheeling & L. E. R. Company v. Carpenter, 218 F. 273 (CCA 6).” Baldwin Rubber Company v. Paine & Williams Company, 107 F.2d 350, 353 (6th Cir. 1939).

*671 Plaintiffs cite the intermediate sales by ARCT-France to Whitin and ARCT, Inc., as further evidence in support of their exhaustion argument, but in the court’s view these sales did not serve to free the machines from the use right restriction any more than if the plaintiffs had purchased directly from ARCT-France. In either case, they would have purchased from a party who had no use rights in the machines, a fact well known to all concerned. Thus the sales remained conditional under the patent laws. Since ARCT-France’s right to manufacture and sell the machines was a contractually-limited one, the use rights were effectively reserved and Chavanoz and DMRC had the right to license the use of the machines separately from their manufacture and sale, Brulotte v. Thys Company, 379 U.S. 29, 85 S.Ct. 176, 13 L.Ed.2d 99 (1964); General Talking Pictures Corporation v. Western Electric Company, 304 U.S. 175, 58 S.Ct. 849, 82 L.Ed. 1273 (1938); In Re Yarn Processing Patent Validity Litigation, 541 F.2d 1127 (5th Cir. 1976); Extractol Process, Ltd. v. Hiram Walker & Sons, Inc., 153 F.2d 264 (7th Cir. 1946).

In the Schwinn case, so heavily relied on by plaintiffs, the Supreme Court extended the per se doctrine to cover restrictions imposed by a vendor on the resale by a distributor of unpatented merchandise as to which the vendor had surrendered title, dominion and control. The plaintiffs have cited no case, however, and the court has found none, extending Schwinn to a restriction imposed by a patentee in the lawful exercise of his patent monopoly rights. 7 Having concluded that the sales by ARCTFrance did not exhaust the patent monopoly, the court declines to apply the Schwinn doctrine to the facts of this case.

The Grant-Back Clause

Covenants in a license requiring the licensee to assign or license any improvements he may make to the patentee, commonly referred to as “grant-backs”, are not as such inherently illegal. Transparent-Wrap Machine Corporation v. Stokes-Smith Company, 329 U.S. 637, 67 S.Ct. 610, 91 L.Ed. 563 (1947). Plaintiffs contend, however, that the contractual obligation of ARCT-France to assign to Chavanoz all rights in improvements in the “patent field” as defined in paragraph 1 of the 1962 agreement “far exceeds the scope and form of grant-back obligation exempted from the per se rule in [Transparent-Wrap]”, and constituted a per se violation of Section 1 of the Sherman Act as well as an unreasonable restraint on trade under that statute.

Because the scope of the improvements required to be granted back by ARCT-France extended substantially beyond the scope of Chavanoz’s original patents, the question here is a close one, but the court has concluded that the grant-back clause did not in this instance offend the antitrust laws. Considerations leading to this determination include the following:

1. The grant-back had no adverse effect on competition in the manufacture of false twist machinery. Neither Chavanoz nor DMRC manufactured machinery, and ARCT-France’s principal competitor in the United States, Leesona, was not affected by this grant-back arrangement between Chavanoz and ARCT-France.

2. ARCT-France was the only manufacturing licensee involved in the grant-back arrangement, and it was free to incorporate its own inventions in its machines without payment of further royalties to Chavanoz. Since all ARCT machines sold in this country included at least two inventions patented initially to Chavanoz in its own right *672 (DX 628), the plaintiff use licensees were under a continuing obligation to pay the level royalty rate established by Chavanoz and DMRC. When new improvements were developed and incorporated in the machines they were made available to plaintiffs without any increase in this fixed royalty rate. Thus the plaintiffs were not harmed by the grant-back arrangement.

3. Invention by ARCT-France was not discouraged by the grant-back clause. Of the twenty-two patents originally in suit here twelve were developed by ARCTFrance and assigned back to Chavanoz. Its research and development efforts obviously were not stifled by the arrangement.

In the early DMRC standard use license form there was incorporated a grant-back clause applicable to DMRC’s use licensees, the plaintiffs in this case, but this provision was deleted in 1961 following an amendment to the Chavanoz-DMRC agreement. The questions raised by the several grant-back clauses which continued to appear in the Chavanoz-ARCT-France agreements will be re-examined under the patent misuse section of this memorandum, but at this point the court is of opinion that the requirement that ARCT-France assign back to Chavanoz all improvements it might make in the “patent field” did not rise to the level of an antitrust violation.

The Price-Fixing Charge

In support of their position on the price-fixing charge plaintiffs argue that the amount of the royalty established by the agreement between Chavanoz and DMRC which the licensees were to pay, and which DMRC consistently maintained was non-negotiable, constituted a part of the sales price of the machines. The fixing of this portion of the price paid by the Throwsters, the plaintiffs contend, constituted a per se violation of Section 1 of the Sherman Act.

Here again we find the plaintiffs relying on the assumption that the sale by ARCTFrance of the machines exhausted the patent monopoly and carried with it an implied license to use the machines without payment of royalties. Since the court has been unable to accept this argument, the charge of price-fixing must be examined in the light of the unquestioned right of a patentee “to exact royalties as high as he can negotiate within the leverage of that monopoly.” Brulotte v. Thys Company, 379 U.S. 29, 33, 85 S.Ct. 176, 179, 13 L.Ed.2d 99 (1964).

Plaintiffs’ reliance on such cases as Ethyl Gasoline Corporation v. United States, 309 U.S. 436, 60 S.Ct. 618, 84 L.Ed. 852 (1940), and United States v. Univis Lens Company, 316 U.S. 241, 62 S.Ct. 1088, 86 L.Ed. 1408 (1942), is misplaced. In each of those cases the patentee had licensed a manufacturer to make and sell the patented product but had reserved the use rights to itself. The patentee had then issued use licenses to wholesalers and retailers, and in each license had set the price at which that party could sell to the next party in the chain. Therefore, on the purported basis of a retained use license, the patentees had set the price paid by the first wholesaler to the manufacturer, by the first retailer to the wholesaler and by the public to the retailer. The holding in these cases is simply that the Sherman Act prohibits the use of a patent monopoly to fix the resale price once the product has passed into the hands of a purchaser from a manufacturing licensee. They in no way impose a restriction on the amount a patentee may set as the purchase price for his invention or the amount he may exact as a royalty for its use. Eastern Venetian Blind Company v. Acme Steel Company, 188 F.2d 247, 253 (4th Cir. 1951).

In Re Yarn Processing Patent Validity Litigation, 541 F.2d 1127 (5th Cir. 1976), another case relied on by plaintiffs, is also distinguishable. In that case the patentee, Leesona, licensed other manufacturers to make and sell machines incorporating its patents but reserved the right to charge the purchasers a production royalty one-third of which was shared with its competitor manufacturers. The Fifth Circuit held that the sales price of the machinery consisted of two elements, the initial price and the royalty payments, and that since the royalty rate was not negotiable by the manufactur *673 ers, this portion of the purchase price of the machinery was fixed. A finding by the district court of violations of Sections 1 and 2 of the Sherman Act was affirmed. The court said:

“A patentee may usually exact whatever royalty it wishes. But, Leesona and Permatwist elected to take a one-third reduction in their royalty income. There is nothing in the patent laws that allows them to decide unilaterally that the machine manufacturers would get the entire benefit of their own royalty reduction. By allocating this benefit, Leesona guaranteed income to the manufacturers and effectively fixed the price of the machinery. The machinery manufacturers who participated in the scheme were protected against free competition and free bargaining in effecting their sales to throwsters.” 541 F.2d at p. 1136.

Although this court is inclined to agree with the plaintiffs that the royalties paid by them to DMRC constituted in effect a part of the purchase price for the machines, the royalties were not shared with a manufacturer but were simply retained by Chavanoz and its licensing agent, DMRC, as the patentee’s reward for its patented inventions. The court concludes that the plaintiffs’ price-fixing charge has not been established.

The Tying Arrangement

Plaintiffs strongly urge that the defendants conspired to impose on them a tying arrangement which constituted a per se violation of Section 1 of the Sherman Act. More specifically, they contend that as a result of the requirement that they take the DMRC use license, patents and unpatented technical information were tied to the machines and non-applicable patents and unpatented technical information were tied to other patents. Although the facts on which plaintiffs’ contentions are based may be seen later to justify a finding of patent misuse, in the court’s view they do not support the conclusion that the tying arrangement violated the antitrust laws.

A tying arrangement is an agreement by a party to sell one product (the tying product) but only on condition that the buyer also purchase a different (or tied) product. Illegality under the tying cases is established when it is shown that the antitrust defendant has sufficient market power with respect to the tying product to restrain free competition in the market for the tied product and a not insubstantial amount of commerce in the tied product is affected by the arrangement. Kentucky Fried Chicken v. Diversified Packaging, 549 F.2d 368 (5th Cir. 1977); Advance Business System & Supply Company v. SCM Corporation, 415 F.2d 55 (4th Cir. 1969); Donlan v. Carvel, 209 F.Supp. 829 (D.Md.1962).

When the tying product is patented sufficient market power to enforce a tie-in is presumed. United States v. Loew’s, Inc., 371 U.S. 38, 83 S.Ct. 97, 9 L.Ed.2d 11 (1962). By the same token when the patent itself is employed as the tying “product” the power of the patentee to require a licensee to purchase a different product is unquestioned. Although the ARCT machines as such were not patented, a rapid growth in their sales was experienced almost from the beginning, and for present purposes the court has assumed that both with respect to the patents and the machines the defendants possessed sufficient economic power in the tying products to enforce the alleged tie-ins at all times. Attention need therefore be focused only on the question of whether a substantial amount of commerce in the tied property was affected by the tying arrangement. This in turn involves identification of the tied product.

Plaintiffs assert that the patents themselves and certain unpatented “technical information” were tied products — that they were tied to the machines — and that plaintiffs were required to take a license under all of the Chavanoz patents (regardless of their applicability to the machines) and the unpatented technical information in order to get the machines. 8

*674 Plaintiffs further complain that when the patents actually applied to the machines are considered as the tying products, this same technical information and many non-applicable patents were tied to them. The tied products are thus identified as the unpatented technical information and certain patents which admittedly were never applicable to any of the machines purchased by the plaintiffs.

The question thus becomes: Was a substantial amount of commerce in these tied products affected by the tying arrangement? The answer must be in the negative. The rationale underlying the rule in the tying arrangement cases is that

“[Tying agreements] deny competitors free access to the market for the tied product, not because the party imposing the tying requirements has a better product or a lower price but because of his power or leverage in another market. At the same time buyers are forced to forego their free choice between competing products.” Northern Pacific R. Company v. United States, 356 U.S. 1, 6, 78 S.Ct. 514, 518, 2 L.Ed.2d 545 (1958).

Such is not the case here. The only serious competitor of the defendants at the time DMRC’s licensing program began was Leesona. Like ARCT, it was in the business of selling yarn texturing machinery embodying its own patents, and except as it was able to sell these machines to the plaintiff Throwsters there was obviously no market among them for its own patents whether adaptable to the ARCT machines or not. Even if Leesona had possessed its own brand of the nebulous “technical information”, as to which the record is silent, it is not readily apparent how it would have been of any use to the purchasers of ARCT machines. It follows that the plaintiffs were not forced to forego their free choice between competing products for there were none.

“. . . Unless a defendant can establish certain narrow affirmative defenses, a finding that the defendant’s conduct falls within the category of per se tying arrangements disposes of the case in the plaintiff’s favor.

“Here, as elsewhere, however, the per se label can sometimes prove misleading. Per se analysis is susceptible to the unwarranted inference that a plaintiff prevails in a tying case merely by finding some way to characterize the defendant’s conduct as a tie . . . To bring a defendant’s conduct within the category of ties that are per se violations of the Sherman Act, however, a plaintiff must go beyond some colorable characterization of the arrangement as fitting this rough definition.

“A plaintiff must show that the challenged arrangement is in fact a tie: that two separate products are involved and that, in addition to complying with the literal terms of the imprecise definition, the seller’s behavior follows the general pattern found unacceptable in the earlier tying cases. To measure an arrangement against that general pattern we must take into account the principal evils of tie-ins: they may foreclose the tying party’s competitors from a segment of the tied product market, and they may deprive the tie’s victims of the advantages of shopping around . . . ” Kentucky Fried Chicken v. Diversified Packaging Corporation, 549 F.2d 368, 375 (5th Cir. 1977).

*675 In summary, although the defendants possessed sufficient market power with respect to the tying products to restrain free competition in the market for the tied products if such market had existed, there was no such market and no commerce in the tied product was affected by the arrangement. Clayton Manufacturing Co. v. Cline, 427 F.Supp. 78 (C.D.Cal.1976). The court therefore fails to find an antitrust violation in the tying arrangement charged here.

Some of plaintiffs charges in connection with the alleged vertical conspiracy will be re-examined later in the section of the memorandum on misuse of patents, but at this point the court concludes that these charges have not been established as antitrust violations.

B. The Horizontal Conspiracy

Plaintiffs’ allegations of conspiratorial conduct of the defendants and their principal competitor, Leesona, which led to the settlement in 1964 of certain patent litigation then pending between Leesona and the defendants will now be considered. Here again we find the facts are not in substantial dispute, but the parties are poles apart in their interpretation of the facts and their contentions as to the legal significance to be given them.

Plaintiffs contend the settlement was the result of a conspiracy between Leesona and the defendants to preserve and perpetuate their established production royalty licensing programs in which the royalties charged their licensees were virtually identical and not subject to reduction; that this constituted price-fixing in violation of Section 1 of the Sherman Act; and that defendants also conspired to monopolize the false twist machinery industry in violation of Section 2 of the Sherman Act. The defendants deny any wrongdoing and with equal fervor contend that the settlement of this “dangerous litigation” was the result of good faith, arms-length bargaining by the adversary parties all of whom were exercising sound, reasonable and lawful business judgment.

FACTUAL BACKGROUND

The agreement between Chavanoz and DMRC for the licensing in the United States of the ARCT machines was signed on December 31, 1957. At that time the only manufacturer of false twist machines in the American market was Leesona which had begun developing a continuous process false twist machine in the early 1950’s and which it began to market in late 1954. This was the Leesona Model 550 or Superloft machine.

Meanwhile, the Permatwist partnership had begun building and marketing attachments for converting conventional textile texturing machines to produce crimped yarns by false twisting. Permatwist filed three patent applications in January, 1954 covering its machine and process which it was marketing under the trademark “Fluflon”.

In December, 1954, Leesona and Permatwist entered into an agreement whereby Leesona acquired the pending Permatwist patent applications which eventually matured into United States Patents Nos. 2,803,105, 2,803,108 and 2,803,109, all of • which were issued on August 20, 1957. In this agreement Leesona acquired the rights to other related inventions and. future improvements as well as Permatwist’s rights under its outstanding license agreements covering the Fluflon machines previously sold. In return Permatwist was to receive a share of all royalties collected by Leesona and a portion of the selling price of Leesona’s false twist machinery. Thereafter Leesona marketed both the Fluflon machinery and its own Superloft machine.

In the inception the Leesona-Permatwist agreement did not require a production royalty to be charged on the Superloft machines, but purchasers were required to take a license under the Permatwist patent applications, and the Fluflon attachments were continued to be sold on a production royalty basis. In early 1957 when Leesona began marketing its Model 511 or “Saaba” attachment for use in post-treating stretch yarns produced by the false twist process, purchasers, although required to sign a *676 license agreement, were not required to pay a continuing production royalty.

When Chavanoz and DMRC began to consider introducing the ARCT machinery into the United States under a continuing production royalty license Leesona’s Superloft machine was the only integrated false twist machine then being sold in the United States. Chavanoz and DMRC recognized that since the Superloft was being sold on a royalty-free basis, it would seriously affect their proposed production royalty program. Recognizing this threat from the Leesona competition DMRC and Chavanoz had inserted in their 1957 agreement a provision that the production royalty to be charged purchasers of the ARCT machinery would be five per cent of the manufacturer’s list price of the raw yarn processed on the machines but that if there were “substantial unlicensed competition” DMRC could adjust the rate to as low as two and one-half per cent, and in view of the continuing sale of the Leesona machines on a royalty-free basis this was the royalty figure charged the DMRC licensees from the beginning.

In an effort to solve this unlicensed competition problem Chavanoz notified Leesona at a time when its dealings with DMRC were still in the negotiation stage that the Superloft machine infringed one or more of the Chavanoz false twist patent applications and suggested that Leesona should take a license under the Chavanoz patents. Leesona declined these suggestions, but Chavanoz never went so far as to threaten Leesona with litigation based upon its alleged infringement of any Chavanoz patent, and in fact the Chavanoz patents have never been asserted against the Leesona machines in this country. 9

The next approach adopted by Chavanoz and DMRC was the institution in November, 1957, by DMRC of an action in the United States District Court for the Eastern District of New York (the “Brooklyn litigation”) in which DMRC undertook to obtain the Leesona false twist patents for itself. The complaint was based on the grant-back provisions in an earlier contract between DMRC and Leesona relating to DMRC’s edge-crimping (“Agilon”) process. Had this suit been successful, DMRC would have been able to bring the Leesona patents under the Chavanoz-DMRC licensing program then being negotiated, and the elimination of this unlicensed competition would have permitted establishment of the DMRC royalty rate at 5%, the maximum rate fixed by the Chavanoz-DMRC agreement.

Leesona eventually won the suit, Deering Milliken Research Corporation v. Leesona Corporation, 201 F.Supp. 776 (E.D.N.Y.1962), aff’d, 315 F.2d 475 (2nd Cir. 1963), but in the meantime its outcome had been rendered irrelevant by another development. Prior to the sale of the first ARCT machine in the United States Leesona instituted a new production royalty program of its own, and this cleared the way for Chavanoz and DMRC to do likewise. Leesona’s new program followed shortly after its development of a high-speed spindle which greatly increased the spindle speed of its false twist machines. Under its new program Leesona required purchasers of its false twist machines incorporating its new high speed spindles to sign a license agreement providing for a production royalty of 6$ per pound on 70 denier yarn. 10 Leesona thereafter applied the same basic royalty schedule to its later model machines.

DMRC received a copy of Leesona’s new license agreement on or before October 10, 1958, which was shortly after the decision to implement the licensing program had been made, and this decision was embodied *677 m a modification of the Leesona-Permatwist agreement on October 17, 1958. Although it had not licensed any purchasers of the ARCT machines at that time, DMRC did not move immediately to bring its royalty rate in line with Leesona’s. 11

Sales of the ARCT machines by Whitin in the United States began in 1959, and on April 1, 1960, Leesona instituted an action in the United States District Court for the Western District of South Carolina against the Judson Mills Division of Cotwool Manufacturing Company (a corporate predecessor of DMI) for infringement of the three patents which it had obtained from the Permatwist applications (the ’105, ’108 and ’109 patents) allegedly resulting from Judson’s use of its ARCT machines (the “Cotwool litigation”). Shortly thereafter there was instituted in the name of Whitin against Leesona in the United States District Court for the District of Massachusetts a suit for a declaratory judgment that the three Leesona false twist patents were invalid, unenforceable and not infringed by the use of ARCT false twist machines (the “Whitin litigation”). Chávanos and DMRC were the guiding hands behind the institution and prosecution of this suit in Massachusetts.

In November, 1961, Leesona also began an arbitration proceeding against Schwarzenbach-Huber, one of the plaintiffs herein, which was licensed both by Leesona and DMRC, in which Leesona attempted to collect royalties under its license by reason of Schwarzenbach-Huber’s operation of its ARCT machines. Under its licensing agreement DMRC was obliged to defend this arbitration proceeding.

Other legal confrontations between DMRC and Leesona occurred in the United States Patent Office where DMRC undertook through interferences with Leesona to gain control of Leesona’s 108 patent and its post-treating Saaba patent No. 2,864,229.

Shortly after the institution of the Cot-wool and Whitin litigations, Warren A. Seem, one of the Permatwist partners, wrote to Armitage of DMRC on December 27, 1960 and suggested that negotiations looking to a settlement of the controversies then pending in “three different arenas” should be undertaken.

“[W]e propose that instead of slugging it out in the public square, we join hands and do something good for the industry as well as ourselves.

“[T]he opponents are now engaged in licensing identical processes requiring endless patent litigation. To us it looks like both sides are bound and determined to destroy each other and harm the industry they both desire to serve.

“However, being quite familiar with both sides, either by personal contact or reputation, we do not have the slightest doubt that all differences can be settled at the conference table and to the mutual advantage of all concerned. As unauthorized intermediaries, we would like to know whether you are willing to give the conference table another try.” (DX 136).

Shortly thereafter meetings were held in early 1961 at which proposals were advanced for settling the Cotwool, Whitin and Brooklyn litigations and the patent interferences. Leesona proposed that DMRC take a Leesona standard manufacturer’s license under the terms of which DMRC would receive one-third of the royalties collected. DMRC proposed that the Cotwool and Whitin suits be dismissed with admissions only of the validity of the Leesona patents. During the first meeting Davis of Leesona and Seem confirmed to Armitage of DMRC that the Leesona royalty on 70 denier yarn was 6$ per pound.

Meanwhile Leesona continued to license other potential competitors in the false twist manufacturing field by sharing one-third of its royalties with them in return for a license to sell false twist machines only to Leesona’s licensees. Unlike normal license arrangements, these competing *678 manufacturers, eventually about thirteen in all, paid Leesona nothing for their freedom from its infringement claims but instead were rewarded by a share of the royalties collected by Leesona. 12

In March of 1961 Leo Soep, representing Chavanoz, Warren Seem, the Permatwist partner, and Robert Conrad, counsel for Leesona, had a conference in London concerning false twist matters, and during the course of their discussions Soep suggested terms upon which the United States litigation might be settled. Reporting on this conference to Armitage, Soep said:

“As you will remember the DMRC/Chavanoz agreement provides for a royalty reduced to one-half its amount as long as there is substantial unlicensed competition.

“Suppose an agreement is reached with Leesona whereby mutual cross-licensing takes place with the promise of non-assertion of clients, then DMRC will be entitled to increase the rate, and we can consider a split of the increased rate between the three of us.

“Conrad is interested in this proposal but mentioned that he would like an overall arrangement with DMRC .

“I do not mind to tie in the Chavanoz problems with the specific DMRC ones if necessary, but I feel that this is not entirely the case. However if it serves your interest, I do not mind.” (PX 1043).

The Soep proposal met with strong objection by Waters of Whitin who could see only a doubling of the DMRC royalty rate while the Leesona rate remained the same thus making it more difficult for him to sell the ARCT machines in the face of the Leesona competition. And, of course, since Whitin did not share in the production royalties in any event, there was nothing in the Soep proposition for Whitin.

Apparently nothing came of the 1961 settlement discussions, and between 1961 and 1963 the record reflects no further formal settlement negotiations. During this period the Cotwool litigation was consolidated with the Whitin litigation in the District of Massachusetts over Leesona’s strong objection, and the Schwarzenbach-Huber arbitration proceeding was stayed from March, 1961, until April, 1963.

On May 21, 1963, Albert Davis, house counsel for Leesona on patent matters, and Soep met in Paris to discuss the possibility of settling both the United States and French litigations, and Soep’s memorandum of the conference made the following day indicates that Davis renewed Leesona’s previous proposal that it grant to Chavanoz and its affiliates a license to manufacture and sell false twist machinery to Leesona’s licensees and that Leesona’s standard license agreement be modified to the extent necessary to make it applicable to users of the ARCT machines. The royalty rate would continue at 6$ per pound for 70 denier yarn and Leesona would pay Chavanoz and its affiliates one-third of the royalties collected from the licensed users of the ARCT machines. Soep further reported:

“Chavanoz having consulted DMRC, has obtained from this company the agreement in principle to sharing royal *679 ties presently received with Leesona. There are three variations possible:

“(a) DMRC reduces its royalties by a certain amount, Chavanoz by a larger amount, and the total sum of this deducted royalty is paid by DMRC and Chavanoz to Leesona in exchange for a hold harmless clause.

“(b) Chavanoz and DMRC pay an annual fixed royalty taken from royalties received from their customers to Leesona against an exchange of a hold harmless clause and

“(c) Chavanoz and DMRC pay a fixed sum in one payment to Leesona against an exchange of a hold harmless clause.

“An intermediate solution has also been endeavored: Chavanoz and Leesona pool their interests and form a non-profit joint association, for the promotion of stretch yarn in the U.S. This organization would do the showing of both type of machinery produced by Leesona and by ARCT so that a reduction of promotion costs would result to the benefit of Leesona and of Chavanoz.

“This solution should, however, be considered very carefully from the angle of antitrust law.” (PX 363).

Following the discussions between Davis and Soep there were discussions between Davis and Whitin’s house counsel, Ward Smith, in May and June of 1963, which led to a conference between the principals on June 18, 1963, at the Algonquin Club in Boston. At this meeting apparently neither side was willing to make any substantial change in its previous negotiation position, and the consequence was that the meeting broke up without much having been accomplished toward settlement.

Prior to the Algonquin conference, however, there had been one significant development on the DMRC side. On June 12, 1963, despite the problems Whitin was having in selling machines because of the pend-ency of the litigation, DMRC, Chavanoz and Whitin agreed to raise the DMRC production royalty rate from 2V2% to 3V2%. This increase brought the DMRC royalty rate in line with Leesona’s with respect to 70 denier nylon yarn since application of the new rate to the long-established list price of $1.71 per pound for the raw yarn resulted in a charge of $.05985 per pound as against Leesona’s 6(f per pound. The new agreement provided that .6% of the royalties to be collected would go directly to DMRC and the remaining 2.9% would go into an escrow fund to be refunded to DMRC’s new licensees in the event the litigation with Leesona terminated adversely to Chavanoz and its affiliates.

Following the Algonquin conference apparently no further settlement discussions occurred between the adversaries until late 1963 when it became known that the Cot-wool and Whitin litigations which had been consolidated would be scheduled for trial in Boston early in 1964. However, the subject of settlement remained alive between Leesona and the Permatwist partners whose interest would be affected by any settlement because of their arrangement with Leesona. In a memorandum of a conference with Permatwist on August 6, 1963, Robert Leeson recorded that

“What we offer [Permatwist] is a good deal on any way of looking at it. 1. Approx. 10% on value of machine export, and 2. One-half of royalty (illegible) and we do everything in our power to sell machines so we can get (illegible) .04 for both of us; and everything to settle so that all competitors charge a royalty.” (PX 353).

In December, 1963, Leesona was able through certain procedural maneuvers to forestall an immediate trial of the case in Boston, and on January 23, 1964, Armitage and Leeson had a further settlement conference. Like its predecessors this conference resulted in an impasse, Armitage maintaining the position that Leesona should accept a lump sum settlement based on the estimated cost of continuing the litigation and Leeson continuing to insist that the ARCT licensees be licensed under the Leesona patents and that production royalties be divided two-thirds to Leesona and one-third to DMRC/Chavanoz. Reporting on this conference Leeson wrote that he told Armitage

*680 . . that he should ask Soep to reconsider his mathematics — there is more at stake than the cost of a suit. If you win, you lose, and if you lose, you lose — because if the patent is broken, there will be no royalty.” (PX 338).

Armitage thereafter reported this conference to Roger Milliken, president and chief executive officer of the Deering Milliken enterprises, and a few days later Milliken conferred with Leeson. This was followed by another conference between Milliken and Leeson in February of 1964, and in consequence of this second conference Armitage prepared and mailed to Leeson a draft of a settlement proposal on March 3, 1964, which incorporated Armitage’s understanding of the conversations between Mil-liken and Leeson.

Meanwhile, on February 28,1964, a Canadian court had rendered a decision upholding the Leesona patents in an action brought against Leesona in Canada by the Scragg Company, a British manufacturer of false twist machinery. There is no evidence that this decision had been brought to the attention of Armitage before he prepared the settlement draft sent to Leeson four days later, and the court is satisfied that this proposed settlement draft was the outgrowth of Milliken’s conferences with Lee-son and was in no way inspired by the Canadian court’s decision in the Scragg case. 13

Negotiations over the specific terms of the agreement continued throughout March, 1964, and on March 20, 1964, Davis of Leesona mailed Armitage a draft settlement agreement which contained the same basic provisions as the agreement finally signed. The parties mutually covenanted not to sue each other’s licensees and DMRC agreed to pay Leesona $150,000 out of future royalties, a figure arrived at after Armitage had given Leeson an estimate of DMRC’s projected royalty income. In a recorded telephone discussion of this draft agreement between Armitage and Davis on March 24,1964, it was agreed in view of its antitrust implications to delete a provision which would have required DMRC to maintain its rate of payments toward the $150,-000 even if DMRC reduced its royalty rate. Armitage explained to Davis that DMRC’s only source of revenue was the royalties and that if either side reduced royalties the other would have to do so. Armitage went on to explain that “there is no reason for us to reduce royalties unless you force us to do so”, and Davis agreed with this. Armitage further stated that he would dislike to have anything in the agreement “that looks like a penalty against reducing royalty rates because I have in mind this antitrust thing.” (PX 222).

Although the final settlement apparently was not agreed upon until some time in April, four settlement agreement documents between defendants and Leesona were signed as of March 31, 1964, the principal one with which the present litigation is concerned being the one that settled all existing litigation between the parties in the United States (PX 223, Tab A).

The agreement is in the form of an exchange of mutual cross-covenants of the parties not to sue each other or customers of the other under certain listed existing patents with respect to presently existing and future machines and under specified existing patents as well as future patents with respect to presently existing machines provided the machines are licensed either by Leesona or Chavanoz. Unlicensed ma *681 chines receive no protection, and in his recorded telephone conversation with Davis Armitage had stated “. . .if they are not licensees we are not interested in what you do to them and you are probably not interested in what we do to them . . . ” As consideration for the settlement DMRC agreed to pay Leesona 10% of the DMRC royalties collected until $150,000 had been paid. All of the litigation pending between the parties in the United States was to be dismissed without prejudice with each side bearing its own costs.

The actual terms of the settlement were not made public, but in a jointly issued press release the out-of-court settlement of all stretch yarn patent litigation previously pending between the parties was announced. The press release contained the statement that the settlement had followed “close on the heels of the February 28,1964 decision” of the Canadian court upholding the validity of Leesona’s stretch yarn patents in Canada (PX 343). Although factually accurate, this statement was misleading for that, as previously stated, the court is satisfied that the Canadian court decision was not a motivating factor in the settlement. Further evidence that it was not is to be found in the fact that instead of bolstering Leesona’s bargaining position as might have been expected, the Canadian decision apparently had no such effect, for it was Leesona and not DMRC which finally retreated from its hard line position in which it had consistently refused to offer a settlement on any basis other than a two-thirds/one-third split of royalties with DMRC with Leesona receiving the larger share.

The settlement agreement itself contained a recital which made it appear to settle more than was actually involved in the pending litigation when it stated that “Chavanoz, DMRC and Whitin allege that certain of the aforementioned Chavanoz patents are infringed by the manufacture, sale or use of certain Leesona yarn processing equipment by Leesona and/or its customers”. The fact is that the possible infringement of Chavanoz’s United States patents by Leesona had never been an issue in any of the litigation and that these patents had never been formally asserted against Leesona.

Certain documents written shortly after the consummation of the settlement agreement shed further light on the intent of the parties with respect to the reasons underlying the settlement. For instance, following a telephone conference with Armitage concerning the wording of the joint press release to be issued a Whitin vice president wrote to Armitage a “personal and confidential” letter dated April 8,1964, in which he stated:

“We grant the desirability of indicating to the world at large that the Leesona and Chavanoz patents are strong and that competition against them from any outside source would be difficult at best.” (PX 344).

A few days later Armitage wrote to a Finnish manufacturer of false twist machines who was seeking a United States representative and explained that as a result of the March 31,1964 settlement agreement the validity of the Leesona patents was undisputed and that “this would put a very considerable burden upon anyone attempting to introduce another false twist machine in the United States.” (PX 533).

Another example of the cooperation between these parties following the settlement was the action of Armitage following his receipt of notice that Turbo Machine Company planned to offer for sale on a royalty-free basis a limited purpose texturing machine of the false twist type. Recognizing that “this development can be harmful to our program”, Armitage requested that a study be made to determine if the Turbo machine infringed any of the Chavanoz patents at the same time observing that the Turbo machine probably infringed Leesona’s patents, and if so, that he “should like to call this to Leesona’s attention for whatever action they may feel advisable to take.” (PX 463).

In summary, the massive volume of evidence offered at the trial and again reviewed in detail post-trial has served to *682 satisfy the court by its substantial preponderance that the dominant purpose of the March 31, 1964 agreement, as reflected in the statements and conduct of the participants both before and after that date and in the terms of the agreement itself, was anti-competitive, that purpose being to preserve and enhance the interdependent royalty programs of Leesona and Chavanoz/DMRC which a trial of the pending litigation might well have destroyed. The court is unable to accept the explanation proffered by DMRC and Chavanoz that they feared the Leesona patents might be upheld, for they had long had the opinion of eminent counsel that these patents were invalid, a judgment which was temporarily vindicated by Judge Atkins in the Florida litigation. 14 The validity of the Chavanoz patents was not jeopardized, for the validity of these patents had not been brought into issue in the litigation. Nor was concern for the sales of ARCT machines a basis for the settlement, the evidence showing that the sales of these machines at the time of the settlement were booming.

A contention advanced by Robert Waters in his testimony that a Leesona victory in the litigation would have put Whitin out of business is not convincing. It is true that a judgment holding the Leesona patents valid would have allowed Leesona’s license program to continue, and if the judgment had also held the ARCT machines to infringe the Leesona patents, it would have been necessary to license them under the Leesona patents. In this event, however, Whitin and ARCT-France would have been entitled to one-third of the production royalties collected by Leesona at least until the Leesona licensing program for its competitor manufacturers was knocked out by the Fifth Circuit in the In Re Yarn Processing case, supra. Leesona doubtless would have offered its standard manufacturer’s license to Whitin, for its first offer of settlement of the Whitin litigation to which it adhered almost to the end contemplated this very arrangement. Thus Whitin for the first time would have become a beneficiary rather than a reluctant and unpaid participant in a production royalty program.

In addition to the foregoing findings of fact with respect to the alleged horizontal conspiracy the court expressly adopts as its own the following proposed findings of fact submitted by the parties.

1. Plaintiffs’ proposed findings of fact on the antitrust issues Nos. 15.1-15.10 inclusive, 15.12-15.27 inclusive, 15.29-15.53; 15.-55-15.60 inclusive, 15.61 except for its last paragraph, 15.62-15.82 inclusive, 15.84 and 15.85; also 15.54 omitting “ARCT-France.”

2. Chavanoz, DMRC and DMI proposed findings of fact on the anti-trust issues under Section II, Nos. 1-25 inclusive, the first sentence in No. 26, 27, 28, 30-34 inclusive, the first sentence of 35, 36, 37 except for its first sentence, and with this addition: “In the case of Gibbs and Smith, they took a Leesona manufacturer’s license in which Leesona paid them to sell to its licensees while there were lucrative side deals in favor of both Madison and Burlington.” (See PX 1249; PX 1250; Tr. Vol. 83, pp. 16,096-103); No. 38, the first sentence only of 39, 42-58 inclusive, 60-63 inclusive, 66, 68, the first sentence only of No. 70, 73, 75, 77, 79-81 inclusive, 82 with the exception of the statement in parenthesis, the first four sentences of No. 83, 84 — 86 inclusive, 88, 94 except for its last two sentences, 95-98 inclusive, 100, the first sentence of 101,105, 117 and 118.

3. ARCT-France’s proposed findings of fact Nos. 35 with the exception of the clause in the first sentence reading “to the point where only one or two sales were made during the entire year 1962”, 36, 37, 39, the first sentence only of No. 40 and 42.

4. ARCT, Inc.’s proposed findings of fact Nos. 43^5 inclusive, 47 except for its last paragraph, 48 except for the last sen *683 tence in the second paragraph and the entire last paragraph, 72, 73, 81-84 inclusive, 86 and 87.

Additional findings of fact will be made later in connection with the individual cases of DMI, ARCT-France and ARCT, Inc.

LEGAL CONCLUSIONS — HORIZONTAL CONSPIRACY

I. Jurisdiction and Venue.

The court has jurisdiction of these actions under 28 U.S.C. §§ 1331, 1332, 1337, 1338, 2201 and 2202 and under 15 U.S.C. §§ 15 and 26. Venue is proper under 28 U.S.C. §§ 1291(c) and 1291(d) and under 15 U.S.C. §§ 15 and 22.

The commerce involved in or affected by the matters in controversy is interstate and foreign commerce within the meaning of Sections 1 and 2 of the Sherman Act, 15 U.S.C. §§ 1 and 2, and the Wilson Tariff Act, 15 U.S.C. § 8.

II. Liability of Defendants.

The application of settled antitrust legal principles to the fact findings detailed above has led the court to conclude that there accompanied the settlement of March 31, 1964, the intent and an implicit agreement to stabilize and maintain production royalty rates and to monopolize the false twist machinery market in the United States; and that the actions pursuant thereto by at least some of the parties constituted violations of Sections 1 and 2 of the Sherman Act. These principles will now be reviewed and their application to the conduct and actions of the separate defendants will be considered.

1. Although settlements of patent litigation are normally as desirable as settlements of other types of litigation, Aro Corporation v. Allied Witan Company, 531 F.2d 1368 (6th Cir. 1976), settlements of such litigation are not sanctioned by the courts when they are attended by anti-competitive results. United States v. New Wrinkle, Inc., 342 U.S. 371, 72 S.Ct. 350, 96 L.Ed. 417 (1952); United States v. Line Material Company, 333 U.S. 287, 68 S.Ct. 550, 92 L.Ed. 701 (1948); Standard Oil Company v. United States, 283 U.S. 163, 51 S.Ct. 421, 75 L.Ed. 926 (1931); Duplan Corporation v. Deering Milliken, Inc., 540 F.2d 1215 (4th Cir. 1976); Westinghouse Electric Corporation v. Bulldog Electric Products Company, 179 F.2d 139 (4th Cir. 1950). 14A

2. Agreements in violation of the antitrust laws may be inferred from the actions and conduct of the parties and need not rest solely on direct testimony. United States v. Container Corporation of America, 393 U.S. 333, 89 S.Ct. 510, 21 L.Ed.2d 526 (1969); United States v. Masonite Corporation, 316 U.S. 265, 62 S.Ct. 1070, 86 L.Ed. 1461 (1942); Interstate Circuit, Inc. v. United States, 306 U.S. 208, 59 S.Ct. 467, 83 L.Ed. 610 (1939); Eastern States Retail Lumber Dealers Association v. United States, 234 U.S. 600, 34 S.Ct. 951, 58 L.Ed. 1490 (1914).

“Acceptance by competitors, without previous agreement, of an invitation to participate in a plan, the necessary consequence of which, if carried out, is restraint of interstate commerce, is sufficient to establish an unlawful conspiracy under the Sherman Act.” Interstate Circuit, Inc. v. United States, supra, 306 U.S. at p. 227, 59 S.Ct. at p. 474.

3. However, proof of agreement, express or implied, is indispensable to the establishment of a conspiracy under the antitrust laws, and evidence of consciously parallel behavior of competitors, while affording some proof of agreement, is not conclusive. Theatre Enterprises, Inc. v. Paramount Film Distributing Corporation, 346 U.S. 537, 74 S.Ct. 257, 98 L.Ed. 273 (1954).

4. Even if alleged antitrust violators do not intend anti-competitive results, a conspiracy between them may yet be proven by a series of actions purposely taken by them which have an anti-competitive result. *684 United States v. Singer Manufacturing Company, 374 U.S. 174, 83 S.Ct. 1773, 10 L.Ed.2d 823 (1963); American Tobacco Company v. United States, 328 U.S. 781, 66 S.Ct. 1125, 90 L.Ed. 1575 (1946).

5. Horizontal price-fixing constitutes a per se violation of Section 1 of the Sherman Act. United States v. Container Corporation of America, supra; Hartford-Empire Company v. United States, 323 U.S. 386, 65 S.Ct. 373, 89 L.Ed. 322 (1945); Standard Oil Company v. United States, 283 U.S. 163, 51 S.Ct. 421, 75 L.Ed. 926 (1931).

6. Action purposely taken to stabilize prices constitutes a per se violation of Section 1 of the Sherman Act. United States v. Socony-Vacuum Oil Company, 310 U.S. 150, 60 S.Ct. 811, 84 L.Ed. 1129 (1940); United States v. Gasoline Retailers Association, 285 F.2d 688 (7th Cir. 1961).

7. An inference of conspiracy may be drawn from the fact that competitors maintain parallel pricing structures which are interdependent. Wall Products Company v. National Gypsum Company, 326 F.Supp. 295 (N.D.Cal.1971).

8. In cases involving patents if an agreement transcends what is necessary to protect the use of the patent or the patent monopoly it may be found to violate the antitrust laws. Standard Sanitary Manufacturing Company v. United States, 226 U.S. 20, 33 S.Ct. 9, 57 L.Ed. 107 (1912).

9. Although an arrangement under which patents are pooled is not per se illegal,

“[i]f combining patent owners effectively dominate an industry, the power to fix and maintain royalties is tantamount to the power to fix prices . . . Where domination exists, a pooling of competing process patents, or an exchange of licenses for the purpose of curtailing the manufacture and supply of ■ an unpatented product, is beyond the privileges . conferred by the patents and constitutes a violation of the Sherman Act. The lawful individual monopolies granted by the patent statutes cannot be unitedly exercised to restrain competition.” Standard Oil Company v. United States, 283 U.S. 163, 174, 51 S.Ct. 421-425, 75 L.Ed. 926 (1931).

10. Patents as exceptions to the general rule against monopolies are affected with a public interest.

“The far-reaching social and economic consequences of a patent, therefore, give the public a paramount interest in seeing that patent monopolies spring from backgrounds free from fraud or other inequitable conduct and that such monopolies are kept within their legitimate scope.” Precision Instrument Manufacturing Company v. Automotive Maintenance Machinery Company, 324 U.S. 806, 816, 65 S.Ct. 993, 998, 89 L.Ed. 1381 (1945).

Authoritative testing of patent validity is encouraged, and the doctrine of licensee estoppel is no longer the law. Lear v. Adkins, 395 U.S. 653, 89 S.Ct. 1902, 23 L.Ed.2d 610 (1969). 15

11. The two elements of monopolization under Section 2 of the Sherman Act are (1) the possession of monopoly power in the relevant market and (2) intent to monopolize. The willful acquisition or maintenance of monopoly power as distinguished from growth or development in consequence of a superior product, business acumen or historic accident is evidence of an intent to monopolize. United States v. Grinnell Corporation, 384 U.S. 563, 86 S.Ct. 1698, 16 L.Ed.2d 778 (1966).

12. Section 2 of the Sherman Act condemns any enterprise which has exercised power to control a defined market if that power is to any substantial extent the result of barriers erected by its own business methods — even though not predatory, immoral or violative of Section 1 of the Act — unless it is shown that the barriers *685 are exclusively the result of superior skills, superior products, natural advantages, business acumen or the like. United States v. United Shoe Machinery Corporation, 110 F.Supp. 295 (D.Mass.1953), aff’d, 347 U.S. 521, 74 S.Ct. 699, 98 L.Ed. 910 (1954).

13. The use of monopoly power, however lawfully acquired, to foreclose competition, to gain a competitive advantage or to destroy a competitor is a violation of Section 2 of the Sherman Act. United States v. Griffith, 334 U.S. 100, 68 S.Ct. 941, 92 L.Ed. 1236 (1948).

14. To support a recovery under either Sections 1 or 2 of the Sherman Act it is necessary for a plaintiff to show that by reason of the violation complained of he has been injured in his business or property. Section 4 of the Clayton Act, 15 U.S.C. § 15.

The task of applying these settled principles to the facts of this case must now be performed.

A. Liability of Chavanoz and DMRC.

As has been previously noted, Chavanoz and DMRC were responsible for the institution and prosecution of the Whitin litigation in Boston. It was their reaction to the Cotwool litigation which Leesona had instituted shortly before in South Carolina. If there is liability, these two parties are in pari delicto, and the cases against them will be considered as one.

The conclusion that the royalty programs of Leesona and Chavanoz/DMRC were interdependent is inescapable. The record contains plenary evidence of this fact and the awareness of it by the parties on both sides. Simply stated, it would not have been economically feasible to market a false twist machine bearing a substantial production royalty in competition with one being sold royalty-free. The natural result was a desire on the part of each side to preserve its own lucrative licensing program.

In order to achieve this mutually-shared objective Leesona, DMRC and Chavanoz entered into the settlement agreement of March 31, 1964. While on its face the agreement would not seem to offend the antitrust laws, when considered in the light of the surrounding circumstances and the actions of the parties both before and after its execution, the conclusion is that the agreement was anti-competitive and a restraint on interstate commerce in violation of Section 1 of the Sherman Act. This conclusion is supported by both the direct and circumstantial evidence previously outlined which in summary form is recapitulated as follows:

1. At the time of the introduction of the ARCT machines into the United States in 1959 the false twist machinery market was completely dominated by Leesona.

2. Since Leesona’s machines were being sold without royalty, it was a matter of concern to Chavanoz and DMRC that the machines on which they hoped to collect royalties would be faced by Leesona’s unlicensed competition.

3. Before the ARCT sales program got underway, however, Leesona instituted its own production royalty program.

4. Thereafter Leesona instituted its manufacturers licensing program with its unusual provision that competing manufacturers of false twist machinery instead of paying royalties to Leesona would in fact receive one-third of the production royalties being collected by Leesona from its licensees but on the condition that the licensed manufacturers sell only to Leesona licensees.

5. The increase in DMRC’s royalty rate brought it in line with Leesona’s rate on 70 denier yarn which constituted about 80% of the false twist production at that time. 16

*686 6. Information as to royalty rates was communicated between the parties as was their recognition of the mutual interdependence of their respective royalty programs.

7. The Leesona patents in suit were known by DMRC and Chavanoz to be weak and, as they had been advised by highly qualified patent counsel, they were confident that these patents could be invalidated. At the same time the validity of the Chavanoz patents was not in issue. Leesona feared a showdown and had always settled its patent litigation short of trial. Insofar as the court is informed, no American court has held these particular patents valid after a trial. 17 Nevertheless, DMRC and Chavanoz were willing to join in a settlement which insured that these patents of dubious validity as well as their own would be immune from attack under the licensee estoppel doctrine which was then the law.

8. The settlement agreement, itself with its cross-covenants not to sue amounted in effect to the cross-licensing of the Chavanoz and Leesona patents by the two patent owners who effectively dominated the false twist industry, and this gave them the power to fix and maintain prices in the form of royalties which they consistently exercised thereafter. 18

9. A misleading press release was prepared by the parties announcing the settlement to the industry.

10. The conduct of the parties following the settlement showed their cooperation in adhering to the program of fixed and stabilized royalty rates and their cooperation in keeping out competition. That they intended the settlement to have this effect is manifest on this record.

The court has also concluded that the settlement agreement of March 31, 1964, resulted in monopolization of the market in the sale and licensing of false twist texturing machinery, processes and technology by Leesona, DMRC and Chavanoz in violation of Section 2 of the Sherman Act. As previously stated, Leesona was the only manufacturer in this market when the ARCT machines came on the scene in 1959, and the Leesona and ARCT machines almost completely dominated the market for more than ten years thereafter. DMRC and Chavanoz controlled the sale of the ARCT machines in the United States, and in combination with Leesona there is no question that they possessed monopoly power in the relevant market. Their intent to monopolize is amply evidenced by their willful acquisition and maintenance of that power, not as the result of superior skills, superior products, natural advantages, business acumen, or the like, but as the result of barriers erected by their own business methods for the purpose of foreclosing competition.

In addition to the foregoing antitrust violations DMRC and Chavanoz by means of their trade-restraining combination with Leesona have knowingly facilitated the combination between Leesona and other competing manufacturers of false twist machines pursuant to licensing agreements which have been held by the Fifth Circuit to constitute per se violations of Sections 1 and 2 of the Sherman Act, In Re Yarn Processing Patent Validity Litigation, 541 F.2d 1127 (1976), and by aiding Leesona in the furtherance of conduct thus held violative of the Sherman Act DMRC and Chavanoz have themselves violated the Act.

Defendants’ arguments have been carefully considered but have been found unpersuasive. It it true that certain pressures were operating on Leesona which made settlement advisable from its standpoint. It had been engaged in expensive litigation for seven years and it was facing formida *687 ble, well-financed adversaries. The contention that the court in which the case was pending was notoriously unfriendly to patents is hardly worthy of mention, but in any event these several “pressures” could have been promptly alleviated by the acceptance by Leesona of the first offer of settlement made by its adversaries: a judgment declaring Leesona’s patents valid but not infringed.

Assuming the validity of the reasons advanced for Leesona’s interest in settlement, these same reasons afforded an additional cause for DMRC and Chavanoz not to settle. But it is said that there were various other pressures operating on these defendants, too, among which were the expense of fighting the suit, the decision of the Canadian court upholding the Leesona patents, the non-profitability of the DMRC licensing program 19 and the burden on sales which the pendency of the litigation created.

The prosecution or defense of litigation is, of course, expensive, but there is nothing in this record to show that for either side the point had been reached where “the game was not worth the candle.” As has already been noted, the court is not persuaded that the outcome of the Canadian case had any appreciable influence on the decision of DMRC and Chavanoz to settle. The profitability of the licensing program was dependent on sales of the ARCT machines, and the fact is that sales were booming at the time of the settlement.

Had Leesona won the suit, defendants argue, it would have been “heaven” for Leesona and “total disaster” for defendants. The court does not envisage that such dire consequences would have attended a Leesona victory. Certainly a judgment limited only to a declaration of the validity of the Leesona patents would have had no such effect, and even if coupled with a finding of infringement of those patents by the ARCT machines, as heretofore suggested, all would not have been lost.

Although defendants’ arguments are not without logical appeal, the definite impression still persists that DMRC and Chavanoz entered into the Whitin litigation confidently expecting to win it and inflict “total disaster” on Leesona. As the case wore on it became more and more apparent to both sides that the results of victory might well be outweighed by the possibility of facing unlicensed competition in the false twist machinery market, a thought epitomized in Leeson’s statement to Armitage: “If you win you lose, and if you lose, you lose — because if the patent is broken, there will be no royalty” (PX 338). This thinking finally prevailed, and the settlement of March 31, 1964 with its trade-restraining, anti-competitive results soon followed.

On the question of the impact these results had on the plaintiffs it is not seriously challenged that unless the plaintiffs in this action signed the DMRC use license and paid the required royalties they could not obtain the ARCT machines which they badly needed in their businesses. The sizable royalty income collected by DMRC in the years following the settlement affords graphic proof of the impact which the program had upon those businesses. 20 The court concludes that as a direct consequence of the violations of Sections 1 and 2 of the Sherman Act by DMRC and Chavanoz the plaintiffs have been injured in their business and property within the purview of 15 U.S.C. § 15. Zenith Radio Corporation v. Hazeltine Research, 395 U.S. 100, 113, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969); Response of Carolina, Inc. v. Leasco Response, Inc., 537 F.2d 1307, 1321 (5th Cir. 1976).

*688 B. Liability of DMI.

Plaintiffs seek to hold Deering Mil-liken, Inc. liable with respect to the Leesona settlement on the theory that its wholly owned subsidiary, DMRC, was at all times acting as its agent or alter ego. DMI strongly resists, contending that the two companies operated at all times as separate corporate entities and that DMI’s involvement in the settlement was limited solely to the action of its president and chief executive officer, Roger Milliken, in arranging for a meeting between Norman Armitage and Robert Leeson in which settlement negotiations could be resumed. The court has concluded that DMI was in fact far more deeply involved and that DMRC’s liability should attach to DMI under agency principles.

DMRC is the successor corporation to a Massachusetts trust, Deering Milliken Research Trust, formed in late 1945 by two New England textile mills which were affiliated with Deering Milliken, Inc. These two mills were later joined by fifteen or more other textile mills which were selling through the Deering Milliken organization, the purpose being to conduct research primarily in the textile field for the benefit of the member companies of the trust. When DMRC was formed in 1955, DMI became the owner of all its issued stock, and it succeeded to the research activities of the research trust and was thereafter operated for the benefit of Deering Milliken, Inc. and its subsidiaries.

DMRC owns no real estate, but it has other fixed assets such as machinery and furniture. Its operating revenue is obtained through two sources, its licensing programs and the research work it does for DMI. As late as 1972 the revenue from its licensing programs had not been sufficient to cover DMRC’s operating expenses. Deficiencies are made up by DMI. DMRC has never declared a dividend, but if it did DMI as its sole stockholder would receive the dividend.

Roger Milliken is the president and chief executive officer of DMI, a very large, privately-held, diversified textile manufacturer engaged in a broad range of commercial textile operations. It employs 22,000 people in sixty-five plants located throughout the United States and Europe.

Mr. Milliken is a man of exceptional ability and stature, and in addition to overseeing the operation of his own huge industrial empire, he holds directorships in several other industrial and financial institutions of national importance.

Despite the demands on his time which his many outside activities involve the impression persists that Mr. Milliken maintains strong personal control over DMI and its subsidiaries, and although he must of necessity rely heavily on the managing officers of the various subsidiaries, major decisions are made by him.

The Leesona settlement certainly was such a major decision, and there is little doubt that Mr. Milliken himself worked out the basic terms of the agreement which was eventually signed. Evidence of this is to be found in Armitage’s letter to Leeson of March 3, 1964 (PX 220), the first paragraph of which read as follows:

“Attached is a draft which I dictated for Roger’s inspection and primarily to be sure that there was no misunderstanding between him and me as to his recollection of his conversation with you. The only point which I did not include was that which I mentioned to you on the telephone this morning, as I felt that it was not of sufficient importance to either of us.”

By contrast, Jerry Cogan, who was DMRC’s executive vice president at the time of the settlement, and later its president, testified that he had nothing to do with the settlement and knew nothing about it until after it had been consummated. He said that he “assumed” that Armitage had had the final say-so in the matter but that he did not know. Cogan said that after he became president of DMRC he learned for the first time that his company was paying royalties to Leesona. There is no doubt that Mr. Milliken oversees his work, Cogan testified, and the person to *689 whom he reports most regularly and keeps fully advised as to what he is doing is Mr. Milliken.

At one time DMRC started research in the chemical field, but when the manufacture of chemicals became a thriving business,- it was transferred by Mr. Milliken from DMRC and became a separate division of DMI.

In addition to the foregoing findings of fact with respect to DMRC’s corporate relationship to DMI the court expressly adopts as its own plaintiffs’ proposed findings of fact on the antitrust issues Nos. 16.1 to 16.12 inclusive.

It is well settled that the fact that a parent corporation owns all of the corporate stock of its subsidiary does not destroy the identity of the latter as a distinct legal entity. However, where one corporation is so organized and controlled and its business is conducted in such manner as to make it merely an agency, instrumentality, adjunct or alter ego of another corporation the separate corporate existence of the parent and subsidiary may not be recognized. 18 Am.Jur., Corporations, § 17; An-not., “Liability of Corporation for Torts of Subsidiary”, 7 A.L.R.3d 1343. In determining whether there has been such domination of the finances, policy and practices of the subsidiary that it has no separate mind, will or existence of its own but is simply a business conduit for the parent corporation various factors may be considered among which are the following:

1. Common stock ownership.

2. Common directors or officers.

3. Financing of the subsidiary by the parent.

4. Incorporation of the subsidiary by the parent.

5. Gross inadequacy of capital for the subsidiary.

6. Payment by the parent of the salaries and other expenses or losses of the subsidiary.

7. The subsidiary has substantially no business except with the parent and no assets except those conveyed to it by the parent.

8. The parent uses the subsidiary’s property as its own.

9. The executives of the subsidiary do not act independently but take their orders from the parent corporation in the latter’s interest.

Bay Sound Transportation Company v. United States, 350 F.Supp. 420, 426 (S.D. Tex.1972), aff’d, 474 F.2d 1397 (5th Cir. 1973).

Most, if not all, of these factors are present here, and the conclusion is that DMRC was so dominated by DMI through its president and chief executive officer, Roger Milliken, in the conduct of its business that it must be held to have been merely an agency and instrumentality of its parent corporation in the transactions which led to the settlement of the Leesona litigation. Accordingly, liability for the resulting antitrust violations found to have been involved in the settlement must attach to DMI. 21

C. Liability of ARCT-France and ARCT, Inc.

Plaintiffs’ efforts to impose liability on ARCT-France and ARCT, Inc. have been directed primarily toward associating these two defendants with the alleged vertical antitrust conspiracy, now rejected by the court as a basis for liability, but as a signatory to the Leesona settlement ARCTFrance is also charged with liability as a member of the conspiracy engendered by that agreement. ARCT, Inc. which did not come into existence until almost two years after the settlement is sought to be charged as a late joiner of the conspiracy. The *690 court has concluded that these charges are not substantiated by a preponderance of the evidence as to either defendant.

In reaching this conclusion the court has considered carefully the relationships between the alleged conspirators and the marked differences between their interests. Normally the participants in an antitrust violation derive their benefits from the same source. Price fixers, for instance, profit from non-competition in pricing. Following the settlement here, however, the sole source of profit for Whitin and ARCT, Inc. continued to be the sale of false twist machinery in open competition with their alleged co-conspirator, Leesona, while Chavanoz and DMRC continued to profit from their production royalty program, freed by the settlement agreement of any threat from Leesona. Likewise the conspiracy of Leesona, DMRC and Chavanoz to monopolize the false twist machinery market did not inure to the benefit of ARCT-France, Whitin or ARCT, Inc. It may in fact have resulted in some competitive disadvantage for them, for as new entrants in the market came on the scene they invariably came under Leesona’s licensing program for manufacturers and thereby qualified to participate in Leesona’s production royalty program to the extent of one-third of the royalties collected, a windfall never enjoyed by Whitin, ARCT-France or ARCT, Inc.

Given this divergence between the interests of the ARCT defendants and their alleged co-conspirators, it is not' surprising that the relationship between ARCTFrance and DMRC was fraught with some friction and antagonism almost from the beginning. As early as March, 1961, we find Crouzet writing to Soep,

“I have told you several times that the presence of DMRC in the arrangements entered into in the United States constitutes a hindrance to us.” (PX 1306)

A few months later Waters on his return from a visit to Europe reported to his superiors that Crouzet was “willing to reimburse DMRC now for out of pocket expenses if DMRC will allow the agreement to be cancelled at this time.”

The relationship between DMRC and Waters and his employers continued to deteriorate, culminating at last in a letter by Armitage to Soep shortly after the formation of ARCT, Inc. in February, 1966, in which he disparaged Waters and questioned the wisdom of the formation of this new company to distribute the ARCT machines in the United States.

When the fact that ARCT and Whitin were to enjoy none of the fruits of the conspiracy (that is, participation in production royalties or freedom from competition in machine sales) is considered in the light of the strained relationship which existed between them and DMRC, the court regards it as highly unlikely that these two companies would have knowingly entered into a conspiracy with DMRC, Chavanoz and Leesona to violate the antitrust laws. It is even more unlikely that Waters would have knowingly permitted his new company, ARCT, Inc., to join the conspiracy when it was incorporated two years later. 22 On the contrary the court is persuaded that in entering into the settlement Whitin and ARCT-France were motivated solely by their desire to sell machines in the United States free from the prospect that their customers might later be faced with the necessity of paying production royalties both to Leesona and DMRC.

The fact that ARCT-France was actually a party to the Leesona settlement agreements is regarded as having little probative force. It was not a party to the United States litigation but was, of course, involved in litigation in France which the settlement encompassed. It was represent *691 ed here by Leo Soep, the ubiquitous “conseil en brevets”, who signed the agreement for ARCT-France as attorney-in-fact. Soep also signed for Chavanoz for whom he had been the chief negotiator. There is no evidence that Henri Crouzet, ARCT-France’s president and chief executive officer, had any part in the negotiations. His limited skill in English makes it unlikely that he had sufficient understanding of the proceedings in the United States to form any intent concerning a conspiracy.

Moreover, Soep’s guilty knowledge and intent, if any, were not necessarily imputable to ARCT-France, for his primary allegiance for years had been to Chavanoz and in this instance he was in a conflict of interest situation. Chavanoz’s sole interest in the settlement was to preserve the production royalty program. On the other hand, ARCT-France’s sole interest was in selling machines, an endeavor for which the DMRC-Chavanoz licensing program posed a constant impediment.

The conclusion is that the requisite intent to violate the antitrust laws on the part of ARCT-France in entering into the Leesona settlement agreement and its actions subsequent thereto have not been made to appear by a preponderance of the evidence. United States v. Singer Manufacturing Company, 374 U.S. 174, 83 S.Ct. 1773, 10 L.Ed.2d 823 (1963); The Duplan Corporation, et al. v. Deering Milliken, Incorporated, et al., 540 F.2d 1215 (4th Cir. 1976). Similarly proof is lacking that ARCT, Inc. knew of the existence of the conspiracy and consciously committed itself to the common scheme with intent to pursue its objectives. Industrial Building Materials, Inc. v. Interchemical Corporation, 437 F.2d 1336 (9th Cir. 1970); United States v. Standard Oil Company, 316 F.2d 884 (7th Cir. 1963); Jones v. United States, 251 F.2d 288 (10th Cir. 1958). As to each of these defendants’ evidence is lacking that there was a common purpose and uniformity of conduct between them and their co-defendants. Theatre Enterprises, Inc. v. Paramount Film Distributing Corporation, 201 F.2d 306 (4th Cir. 1953), aff’d, 346 U.S. 537, 74 S.Ct. 257, 98 L.Ed. 273 (1954). It follows that ARCTFrance and ARCT, Inc. are not liable to the plaintiffs as participants in the horizontal conspiracy.

III. Statute of Limitations.

As an affirmative defense to the antitrust charges growing out of the Leesona settlement DMRC, Chavanoz and DMI have pleaded the bar of the four-year statute of limitations provided for in Section 4B of the Clayton Act, 15 U.S.C. § 15b, which reads as follows:

“Any action to enforce any cause of action under sections 15, 15a, or 15c of this title shall be forever barred unless commenced within four years after the cause of action accrued.”

Defendants contend that the settlement of March 31, 1964 received wide press coverage at the time and was well known throughout the industry. Also well known was the fact that DMRC had obtained the opinion of counsel prior to the settlement that Leesona’s false twist patents should be invalidated. In these circumstances, defendants argue, any cause of action based upon the settlement (which was actually consummated sometime in April of 1964) was barred after April, 1968. Since the earliest pleading here involved which asserted a claim regarding the 1964 settlement was Duplan’s complaint filed in November of 1969, defendants assert that that action and all others which were filed later making the same claim are barred by the four-year statute of limitations.

Plaintiffs on the other hand contend that defendants committed overt acts in pursuance of the conspiracy growing out of the Leesona settlement which continued down through the date of the filing of the Duplan suit including the continued unlawful collection of royalties and that a new cause of action accrued upon the comjnission of each such overt act. Plaintiffs argue, therefore, that they are entitled to recover all damages attributable to these unlawful acts which occurred during the four years prior to the institution of suit.

*692 While the defendants’ position is not without some supporting authority, 23 decision here must be in conformity with the principle stated by the Supreme Court in Zenith Radio Corporation v. Hazeltine Research, Inc., 401 U.S. 321, 338, 91 S.Ct. 795, 806, 28 L.Ed.2d 77 (1971):

“The basic rule is that damages are recoverable under the federal antitrust acts only if suit therefor is ‘commenced within four years after the cause of action accrued,’ 15 U.S.C. § 15b, plus any additional number of years during which the statute of limitations was tolled. Generally, a cause of action accrues and the statute begins to run when a defendant commits an act that injures a plaintiff’s business. (Citing cases.) This much is plain from the treble-damage statute itself. 15 U.S.C. § 15.”

The Supreme Court had applied this general rule three years earlier in Hanover Shoe, Inc. v. United Shoe Machinery Corporation, 392 U.S. 481, 88 S.Ct. 2224, 20 L.Ed.2d 1231 (1968). In that case a shoe manufacturer sued the manufacturer of shoe machinery in 1955 alleging a violation of Section 2 of the Sherman Act dating back to 1912 when defendant instituted its policy of leasing rather than selling its shoe machinery to the plaintiff and other shoe manufacturers. In the Supreme Court the defendant argued that because the earliest impact of its lease only policy occurred in 1912 the plaintiff’s cause of action arose during that year and was therefore barred by the applicable statute of limitations. In rejecting this argument the Supreme Court said:

“We are not dealing with a violation which, if it occurs at all, must occur within some specific and limited time span. . . Rather, we are dealing with conduct which constituted a continuing violation of the Sherman Act and which inflicted continuing and accumulating harm on Hanover. Although Hanover could have sued in 1912 for the injury then being inflicted, it was equally entitled to sue in 1955.” Id., Footnote 15, 392 U.S. at 502, 88 S.Ct. at 2236.

These principles have been consistently followed in this circuit, Charlotte Telecasters, Inc. v. Jefferson-Pilot, 546 F.2d 570 (4th Cir. 1976); Blackwelder v. Millman, 522 F.2d 766 (4th Cir. 1975); Railing v. United Mine Workers, 445 F.2d 353 (4th Cir. 1971), and in other circuits, Twin City Sportservice, Inc. v. Charles O. Finley & Company, 512 F.2d 1264 (9th Cir. 1975); Baker v. F & F Investment, 420 F.2d 1191 (7th Cir. 1970). In a decision rendered since the trial of the case at bar the Fifth Circuit on facts quite similar to those involved here has held that an antitrust cause of action continues to accrue so long as a defendant continues to accept benefits under or assert the validity of a contract entered into in violation of the antitrust laws. Imperial Point Colonnades Condominium v. Mangurian, 549 F.2d 1029 (5th Cir. 1977). 23A District courts have reached the same result. Schokbeton Products Corporation v. Exposaic Industries, Inc., 308 F.Supp. 1366 (N.D.Ga.1969).

In this case DMRC/Chavanoz and Leesona continued to maintain their respective licensing programs and to collect fixed production royalties from the plaintiffs up to and after the filing by plaintiffs of their antitrust actions herein, and these and other overt acts committed by the defendants within the four-year period preceding the institution of the actions created new causes of action in favor of the plaintiffs which are not time-barred.

Additionally, defendants’ plea of the statute of limitations must be rejected on the theory most recently recognized in this circuit in Charlotte Telecasters, supra, *693 that a cause of action does not accrue for purposes of the four-year statute of limitations as to future damages which are speculative and essentially unprovable. Because of the uncertainties in the false twist yarn industry in 1964 and 1965 and the enormous growth and change in character of false twist yarn consumption which occurred within the four years next preceding the filing of these suits it would have been impossible for the plaintiffs to prove with any reasonable degree of accuracy their future damages had these actions been instituted within four years from the 1964 settlement.

The defendants’ plea of the statute of limitations is rejected with respect to all damages which plaintiffs may show have been sustained within four years preceding their suits resulting solely from acts of the defendants committed during the four-year period.

II.

THE PATENT MISUSE ISSUES

In this section of the memorandum it will be assumed that no antitrust violations have been shown and that the Chavanoz patents remaining in the suit are valid. 24 The question for decision is whether the DMRC licensing program as implemented through the use of its standard licensing agreement violated the doctrine of misuse of patents. 25

The Licensing Agreement

The essential provisions of the standard DMRC licensing agreement are set forth on pages 14-17 supra. Briefly it purports to grant to the licensee a non-exclusive and non-transferrable right to use the “FT Processes” and “FT Machines”. “FT Processes” are defined in the preamble as “technical information relating to processes for the manufacture of crimped synthetic yarns based upon the application of a false twist.” “FT Machines” are defined as “certain inventions and technical information relating to . . . devices for the manufacture of crimped synthetic yarns based upon the application of a false twist.” The “inventions” were said to be described in United States patents and patent applications and to be listed in Appendix A attached to the license. In practice this Appendix A was rarely, if ever, attached. As previously noted, just what the term “technical information” encompassed is not made clear by the licensing agreement (see Footnote 8, page 673, supra ). 26

In the spring and summer of 1961 one of the Throwsters through its attorney, one Charles A. McClure of Philadelphia, engaged in an extended exchange of correspondence with Armitage of DMRC (PX 215 and PX 410) in an effort to obtain more definite information as to what the license covered and how the royalty payment related to the various items in the patent package including “technical information”. In his letter of April 17, 1961, McClure wrote:

*694 “[I]t appears to be the intention to collect royalties regardless of whether the conversion [of crimped yarn] involves practice of an invention covered by a licensed patent or patent application. In my opinion, the running royalty should be based only upon the latter, especially in view of the provision for a minimum annual royalty, which itself should be ample recompense for whatever unpatentable technical information might be received by my client from your company.”

In his reply dated April 22, 1961, Armitage said he could not agree. On the contrary, he continued,

“The machine itself contains structural and functional elements, some of which are subject of patents and patent applications and some of which are included in the technical information which is made available to the purchaser of the machine, and this also applies to the process under which the licensee is licensed. As far as the machine is concerned, ARCT, the licensed manufacturer, is not permitted to make available such embodiments of the technical information except to licensees for their use. For this reason the use licensee may well consider that the cost to him of a machine with this right to use it is the original purchase price plus the continuing royalty. Your suggestion that the minimum annual royalty would be ample recompense for technical information is not a realistic one, because this would amount to only $5,000 total for the minimum period of five years. This is not a minimum for each machine but for all the machines which the licensee buys, and it would not begin to pay what it has already cost us to develop only the information upon the effect of variables in machine settings upon yarn properties which we have had to carry out for the American yarns which our licensees will be using. As a matter of fact, you will realize from our own correspondence that just the cost of negotiating this agreement will absorb a generous fraction of the minimum!

The royalty rate, at first two and one-half per cent and later three and one-half per cent “of the manufacturer’s list price of the raw yarn . . . which is converted to crimped yarn by licensee according to the FT process or FT machines”, remained constant regardless of the nonapplicability of some of the Chavanoz patents in the package to the machines. DMRC knew, or should have known from the beginning of the licensing program that many of the patents included in the package were not applicable to any ARCT machine. 27

The original license form contained a grant-back clause requiring purchasers of ARCT machines to grant any new patents and improvements which they might acquire to DMRC or its designee. This provision was eliminated in 1961, but the grant-back provisions in the Chavanoz-ARCTFrance agreements have remained in effect.

The DMRC license agreement gave the licensee the right to terminate the license after five years upon sixty days’ written notice and on this further condition:

“In the event of termination or cancellation of this agreement by operation of paragraphs 10, 11 or 12 hereof, licensee agrees to cease using the FT process and FT machines for the manufacture of crimped yarn according to inventions which are the subject of the present agreement . . . ”

As illustrated in the exchange of correspondence with McClure, DMRC consistent *695 ly took the position that the principal provisions of the license agreement were non-negotiable, and when pressed for explanations or changes it invariably pointed to its agreement with Chavanoz as controlling and limiting its freedom to negotiate. The result was that a Throwster who wanted an ARCT machine took the license in the form tendered him or he did not get the machine.

In addition to the foregoing findings of fact with respect to patent misuse the court adopts as its own the following proposed findings of fact submitted by the parties:

1. Plaintiffs’ proposed findings on the antitrust and patent misuse issues Nos. 14-64,14.66-69 inclusive, 14.72,14.76,14.77, the first paragraph only of 14.78, 14.80-14.82 inclusive, 14.84 except for the last clause in the second paragraph, 14.86 — 14.89 inclusive, 14.98 except for the last five lines, 14.99, 14.103 except for the parenthetical statement in the first paragraph, 14.108, 14.109 and 14.112.

2. Defendants’ proposed findings on the antitrust and patent misuse issues, Section I, Nos. 125-142 inclusive, 152-155 inclusive and the first two sentences of 156.

Contentions of the Parties

On the foregoing facts plaintiffs contend that in addition to their proof of antitrust violations by the defendants, 28 the defendants have been guilty of various other acts of patent misuse including:

1. Compulsory package licensing of patents many of which were known not to be applicable to the ARCT machines.

2. Tying unpatented “technical information” and non-applicable patents to other patents. 29

3. The collection of royalties at a level rate regardless of applicability of patents to the machines and without regard to the expiration of patents.

4. The enforcement by Chavanoz of the grant-back provisions in its contracts with ARCT-France.

5. The alleged requirement that plaintiffs had to cease using the machines upon termination of the license agreement.

6. The alleged requirement that royalties continue after all patents have expired.

7. The assertion by defendants in these actions of patents known to be invalid or not infringed.

8. The alleged deceptive practices of the defendants before the United States Patent Office.

The defendants on the other hand deny any misuse of the Chavanoz patents and at the same time invoke the purge doctrine in bar of any misuse charge that might be established. Defendants further contend that the Chavanoz patents were “blocking” or “complementary” patents which constituted a “single product” thus negating the plaintiffs’ claims of unlawful package licensing. Finally, defendants contend that if plaintiffs were not satisfied with the licensing arrangement tendered them it was their duty to ask for something else; that they never did so; and that they should not now be heard to complain of an arrangement which defendants claim “made them millionaires.”

Conclusions on the Misuse Issue

The court is of opinion that there is merit to some, but not all, of plaintiffs’ contentions, and that when the totality of defendants’ conduct in respect of their licensing program is considered they are seen to have engaged in deceptive business practices which add up to misuse of the patents which has not been purged. Defendants’ *696 actions for patent infringement must therefore fail. Compton v. Metal Products, Inc., 453 F.2d 38 (4th Cir. 1971); cert. denied, 406 U.S. 968, 92 S.Ct. 2414, 32 L.Ed.2d 667 (1972).

Patent misuse is based on the principle that a party seeking to enforce a patent who comes to court with “unclean hands” or who is guilty of “inequitable conduct” should not be afforded relief. It occurs when there is an attempt to extend the patent beyond the scope of its claims. Morton Salt Company v. G. S. Suppiger Company, 314 U.S. 488, 62 S.Ct. 402, 86 L.Ed. 363 (1942). The manner in which the monopoly of the patent is sought to be extended is not material.

“In construing and applying the patent law so as to give effect to the public policy which limits the granted monopoly strictly to the terms of the statutory grant . . . the particular form or method by which the monopoly is sought to be extended is immaterial.” United States v. Univis Lens Company, 316 U.S. 241, 251-2, 62 S.Ct. 1088, 1094, 86 L.Ed. 1408 (1942).

The mere inclusion of two or more patents in a single license agreement does not of itself constitute patent misuse where the parties mutually agree that a group of patents are to form the subject matter of the license agreement. Such voluntary package licensing offends neither the patent nor the antitrust laws. United States v. Paramount Pictures, 334 U.S. 131, 68 S.Ct. 915, 92 L.Ed. 1260 (1948). Where the patent owner refuses to grant a license under less than all of his patents, however, or requires the licensee to accept a license under unwanted or inapplicable patents in order to obtain the use of desired patents, the practice is condemned under the patent laws as mandatory or coercive package licensing. American Securit Company v. Shatterproof Glass Corporation, 268 F.2d 769 (3rd Cir. 1959), cert. denied, 361 U.S. 902, 80 S.Ct. 210, 4 L.Ed.2d 157 (1959).

The court has concluded that DMRC’s license plan falls in this latter category. In order to obtain the ARCT machines the plaintiffs were obliged to take the DMRC standard license agreement in the form offered and pay the level royalty rate charged, and as graphically illustrated in the negotiations between Attorney McClure and Armitage, it was futile for a machine purchaser to undertake to obtain any substantial variance in the terms of the agreement. 30 Others who undertook to obtain more favorable terms met with the same results. 31

“The policy of plaintiff to grant no license unless it is under all its patents for a fixed royalty, is one of unlawful coercion contrary to public policy as announced by the courts. And, it cannot be said by plaintiff in answer that its practice of group or package-licensing was designed merely to suit its business convenience. Convenience to the patent owner can never justify an unlawful use of the patent monopoly. In this field, a reward granted to an inventor is important, only insofar as it serves to release to the public the product of his genius. That service ceases to be rendered when the reward no longer can be found to relate to the quality of each patent. In such event one patent cannot draw strength from the others and the monopoly of each thus be extended.

* * * * * *

*697 “I thus find no significant difference between, on the one hand, conditioning the granting of a license under one patent upon the acceptance of another for an additional consideration, a practice the Supreme Court has condemned, and, on the other, as here, granting a license only under all patents at a fixed rate, or under any one patent at the same rate, which is, in effect, equivalent to apportioning the fixed rate amongst all the patents and then requiring the applicant to take a license under all if he wishes to have a license under one. In either case plaintiff’s policy is inexorable.” American Securit Company v. Shatterproof Glass Corporation, 154 F.Supp. 890, 895 (D.Del. 1957), aff’d, 268 F.2d 769 (3rd Cir. 1959), cert. denied, 361 U.S. 902, 80 S.Ct. 210, 4 L.Ed.2d 157 (1959). (Noted in 44 U.Va. L.R. 727.)

The court is also of opinion that the tying of the unpatented technical information to the patent package was an act of patent misuse. In his correspondence with McClure Armitage made it clear that a substantial portion of the royalty payment was attributable to this nebulous item. Included in the package also were process patents which were capable of being practiced on machines of other manufacturers as well as those made by ARCT. At the same time the ARCT machines could be and were operated extensively without infringing the defendants’ process patents. The requirements that these patents be included in the package was also an unlawful tie.

“The patent monopoly of one invention may no more be enlarged for the exploitation of a monopoly of another . . . than for the exploitation of an unpatented article . . . or for the exploitation or promotion of a business not embraced within the patent.” Ethyl Gasoline Corporation v. United States, 309 U.S. 436, 459, 60 S.Ct. 618, 626, 84 L.Ed. 852 (1940).

As previously noted, the Chavanoz patents were available to the plaintiffs only on an all or none basis. Although made in the context of an antitrust case, the following statement from Advance Business Systems & Supply Company v. SCM Corporation, 415 F.2d 55 (4th Cir. 1969), is equally applicable here:

“Tie-ins are non-coercive, and therefore legal, only if the components are separately available to the customer on a basis as favorable as the tie-in arrangement.” Id. at p. 62.

Furthermore, the misuse of a patent does not depend on a showing that the improper tying provision of the license may have a substantial effect on competition as is required to establish liability under Section 3 of the Clayton Act, 15 U.S.C. § 14. As a matter of patent law the inclusion of tying provisions in a mandatory package patent license constitutes a misuse of patents absent some showing justifying the practice such as business convenience or necessity. Automatic Radio Manufacturing Company v. Hazeltine Research, Inc., 339 U.S. 827, 70 S.Ct. 894, 94 L.Ed. 1312 (1950); Dehydrating Process Company v. A. O. Smith Corporation, 292 F.2d 653 (1st Cir. 1961). See Standard Oil Company v. United States, 283 U.S. 163, 51 S.Ct. 421, 75 L.Ed. 926 (1931). Even then, when the business necessity terminates, the licensee’s rights are re-activated. United States v. Jerrold Electronics Corporation, 187 F.Supp. 545 (E.D.Pa.1960), aff’d per curiam, 365 U.S. 567, 81 S.Ct. 755, 5 L.Ed.2d 806 (1961). And a patentee who sells a patented item only in conjunction with some other item “raises serious suspicions of tying behavior and misuse” and “bear[s] a heavy burden in overcoming this suspicion and in bringing [himself] within one of the [recognized] justifications.” Rex Chainbelt, Inc. v. Harco Products, Inc., 512 F.2d 993, 1002-3 (9th Cir. 1975). See further, Frost, “Tying Clauses and Package Licensing”, 28 U.Pittsburgh L.R. 207 (1966).

In addition to constituting coercion to accept unlawful tying arrangements incorporated in package licenses, American Securit Company v. Shatterproof Glass Corporation, 154 F.Supp. 890 (D.Del.1957), the requirement that the royalty rate remain at a constant level also raises the question of *698 the legality of collecting royalties attributable to patents which expire while the license agreement is in effect. In Brulotte v. Thys Company, 379 U.S. 29, 85 S.Ct. 176, 13 L.Ed.2d 99 (1964), the Supreme Court struck down a patentee’s attempt to collect royalties after all the patents in the package had expired but left open the question of whether the continued collection of the level royalty after the expiration of only part of the patents would constitute patent misuse. 379 U.S. at p. 32, Footnote 5, 85 S.Ct. 176.

On this question the circuits are not in harmony. The Sixth Circuit in Rocform Corporation v. Acitelli-Standard Concrete Wall, Inc., 367 F.2d 678 (6th Cir. 1966), found that the level royalty charged for the use of a number of patents constituted an unlawful effort to collect royalties on an expired patent. The Tenth Circuit, adhering to its decision in an earlier case, declined to follow this in Well Surveys, Inc. v. Perfo-Log, Inc., 396 F.2d 15, 18 (10th Cir. 1968). 32

The DMRC license provided that it should continue “for a period until expiration of the last patent to issue in the United States”, and this court finds itself in agreement with the Third Circuit in American Securit Company v. Shatterproof Glass Corporation, 268 F.2d 769 (3rd Cir. 1959), cert. denied, 361 U.S. 902, 80 S.Ct. 210, 4 L.Ed.2d 157 (1959), when it said:

“We conclude also, and quite apart from all of the foregoing, that paragraph 8(a) of Securit’s standard licensing agreement which provides that that agreement shall continue ‘in full force and effect to the expiration of the last to expire of any’ of Securit’s patents set out in ‘Schedule A’ constitutes a patent misuse for it extends the payment of royalties of patents under patents which may expire to the expiration date of that patent most recently granted to Securit.” 268 F.2d at p. 777.

This is but a logical extension of the rule thereafter adopted by the Supreme Court in Brulotte.

“But the theory of the [Brulotte ] case, standing as it does for the proposition that extending payments beyond the expiration of a patent is a means of converting a seventeen-year monopoly into a longer one (time leverage), seems logically extendable to the proposition that a similar time extension is involved if the payment is not proportionately reduced when one of several patents expires. To take an extreme example, if Brulotte, rather than waiting until all the patents embodied in the hop-picking machine had expired completely to discontinue his payments, had reduced his payments by one-seventh when the first of the seven patents had expired, why would this not also be permissible by application of the theory of the case? Bowman, “Patent and Antitrust Law”, p. 234 (1975).

In the case at bar the expiration of the first applicable Chavanoz patent did not occur until June, 1976, after the trial of the case was already underway. The court has no doubt, however, that had it expired during the time the plaintiffs were still paying royalties and had they undertaken to adjust the royalty rate proportionately, DMRC would have reacted promptly and firmly. 33

*699 The defendants argue, however, that each of the patents actually employed on the ARCT machines was worth the full amount of the royalty charged and that the inclusion of additional patents without additional charge harmed no one. 34 A similar argument was answered most effectively by Judge Dorsey Watkins in Technograph Printed Circuit, Ltd. v. Bendix Aviation Corporation, 218 F.Supp. 1 (D.Md.1963), aff’d per curiam on other grounds, 327 F.2d 497 (4th Cir. 1964), 34A when he said:

“Plaintiffs contend that the royalty rate in fact charged for the use of a single patent is fair and reasonable, and that to permit a licensee to select as many additional ones as he wishes, at the price of one, cannot be coercive. If it were true that the royalty for the use of a single patent were fair and reasonable, there would be great strength to plaintiffs’ argument that the owner of multiple patents need not license each of them at less than a fair and reasonable rate, in order not to be charged with misuse.

“The evidence in the case does not satisfactorily establish whether or not the royalty charged would be reasonable for .a single patent. . . . Moreover, the willingness to throw in any number of patents for the same rate (the only consideration apparently being the estoppel of a licensee to contest the validity of a patent under which he is licensed) is suspect. Altruism is a virtue not usually ascribed to licensors.” Id. 49.

The court has the unmistakable impression that the package license plan employed by defendants was designed to require payment of royalties by the plaintiffs on patents not used on the machines they purchased and to shield from attack under the law as it then existed many weak and unwanted patents. This constituted patent misuse. United States v. General Electric Company, 82 F.Supp. 753, 859 (D.N.J. 1949). 35

We come now to consider the effect of the grant-back clauses in the 1954 and 1962 agreements between Chavanoz and ARCT-France as related to plaintiffs’ charges of patent misuse. In a previous section concerned with plaintiffs’ vertical antitrust conspiracy charges (pp. 671-672, supra) it was concluded that the plaintiffs were not harmed by these grant-back provisions, but in the patent misuse context it is not necessary that a patent licensee show that he has sustained damage by reason of the misuse.

“It is the adverse effect upon the public interest of a successful infringement suit, in conjunction with the patentee’s course of conduct, which disqualifies him to maintain the suit, regardless of whether the particular defendant has suffered from the misuse of the patent.” Morton Salt Company v. Suppiger, 314 U.S. 488, 494, 62 S.Ct. 402, 406, 86 L.Ed. 363 (1942).

The court has concluded that the scope of ARCT’s “improvements” and inventions required to be assigned to Chava *700 noz extended far beyond the scope of Chavanoz’s basic patent (United States Patent 2,741,893, the so-called “bathtub patent”) and the effect was to extend unlawfully its monopoly and thus result in patent misuse.

Prosecution of patent No. 2,741,893 in the United States Patent Office was marked by several rejections before the final allowance of three claims, central to each of which was the claim that the false twisted yarn as it passed through the machine had the twist arrested within the heating medium so that it did not run back to the supply package feed roll. Had the grant-back clauses limited the scope of the improvements required to be assigned to Chavanoz by ARCTFrance to improvements in the alleged twist arresting feature, in all probability they would have satisfied the test in Transparent-Wrap Machinery Company v. Stokes & Smith Company, 329 U.S. 637, 67 S.Ct. 610, 91 L.Ed. 563 (1947). 36 The 1954 and 1962 agreements, however, were not so limited. As shown by the following provision in the 1954 agreement the breadth and scope of the technology required to be assigned to Chavanoz by ARCT-France in the false twist field was virtually unlimited:

“4. . . . [T]he rights of industrial property are reserved exclusively for [Chavanoz] including those attached to the new models. Any application for a patent that could concern such new models must be made by [Chavanoz] in its name and at its expense. ARCT will take the necessary steps to inform [Chavanoz] promptly of any improvements so as to allow it to insure adequate protection of which [Chavanoz] is the sole judge.” (Emphasis supplied.)

Equally broad was the provision in the 1962 agreement which required ARCTFrance to assign to Chavanoz all improvements “in the field of the patents” — defined as “the field of the manufacture of crimped textile yarns, either natural, artificial or synthetic, curled through the application of a false twist which is heat set, and in the yarns which may be obtained from said crimped or curled yarns through an additional treatment or shaping.” 37

In view of these provisions and the fact that it was ARCT-France, not Chavanoz, that was in the machinery manufacturing business it is not surprising that of the patents originally in suit here nine of the sixteen apparatus patents and three of the six process patents originated with ARCTFrance and were acquired by Chavanoz by grant-back. No one of these patents was concerned with a mechanism for “arresting the twist within the heating medium.” 38 That the scope of the monopoly granted in Chavanoz’s own patents was thereby substantially enhanced and extended seems undeniable. This constituted patent misuse. 39

If it were true, as plaintiffs earnestly contend, that upon termination of the DMRC standard licensing agreement licensees were required by the terms of its paragraph 13 (p. 667, supra) to cease using the ARCT machines altogether, or that their obligations to pay royalties were to contin *701 ue notwithstanding the expiration of all patents applicable to the machines, then clearly there would be patent misuse under the rule in Brulotte v. Thys Company, 379 U.S. 29, 85 S.Ct. 176, 13 L.Ed.2d 99 (1964). However, the court is unable to accept plaintiffs’ interpretation of the agreement. The requirement that licensees cease using the FT machines upon termination of the agreement is qualified by the clause, “according to inventions which are the subject of the present agreement.” A fair interpretation of this language, it seems to the court, is the one contended for by the defendants, that is, that upon termination a licensee could no longer use a machine which contained valid, unexpired patents. Nor does a fair reading of the agreement require payment of royalties after all applicable patents on the machine have expired. Although there are isolated instances in which defendants’ witnesses indicated a contrary interpretation of the agreement by DMRC, read as a whole their testimony supports the conclusion the court has reached (Tr. 20, pp. 4105 et seq.).

The plaintiffs have argued that the assertion by defendants of twenty-two patents against them in this litigation when defendants knew or should have known that some of the patents were invalid and at least twelve of them were uninfringed constitutes an independent basis for finding patent misuse or at least an abuse of the patent laws. In adopting defendants’ proposed findings of fact Nos. 1-125 to 1-134, inclusive, however, the court has in effect found that except with respect to patent No. 3,232,037 (the “ ’037 patent”) this argument is without merit. 40 Briefly summarized, these findings show that out of an abundance of caution DMRC initially asserted the twenty-two patents by way of complaint or compulsory counterclaims at a time when no discovery had been conducted; that in most of its pleadings DMRC only alleged infringement of “some or all of the FT patents without prejudice”; that plaintiffs successfully resisted DMRC’s motion for leave to drop some of the patents in the early stages of the litigation; that as discovery revealed instances of non-infringement DMRC brought the information to the court’s attention; and that “[t]he course of conduct was nothing more than an orderly and rational manner of conducting a complex, multi-party litigation.”

The court is inclined to agree, and it finds no patent misuse by reason of the wrongful assertion of any patent other than the ’037 patent. While an attempt to assert a known invalid patent might constitute an antitrust violation in addition to patent misuse, Kellogg Company v. National Biscuit Company, 71 F.2d 662 (2nd Cir. 1934), it is also true that

“Patents would be of little value if infringers of them could not be notified of the consequences of infringement or proceeded against in the courts. Such action considered by itself cannot be said to be illegal. Patent rights, it is true, may be asserted in malicious prosecutions as other rights, or asserted rights, may be. But this is not an action for malicious prosecution. It is an action under the Sherman Antitrust Act for the violation of the provisions of that act, seeking treble damages.” Virtue v. Creamery Package Manufacturing Company, 227 U.S. 8, 37-38, 33 S.Ct. 202, 208, 57 L.Ed. 393 (1913).

With respect to plaintiffs’ claim that defendants through the use of decep *702 tive practices before the Patent Office abused the patent laws the court has adopted defendants’ proposed findings of fact Nos. 1-135 to 1-142, inclusive. Briefly summarized, these findings, while conceding that Judge Hemphill held two of the Chavanoz patents (Nos. 3,137,119 and 3,382,656) to be invalid under 35 U.S.C. § 102(d) because of the failure of the inventor to disclose to the Patent Office that French patents had been applied for more than a year earlier, fail to establish fraud which infect the other patents in the Chavanoz package but at the most establish only negligence or simple human error. It is well settled, of course, that proof of fraud must be clear and convincing and that one alleging fraud has a heavy burden of persuasion. Krenzer v. Stoffel, 551 F.2d 1214 (Oust. & Pat.App. 1977); North Carolina v. Charles Pfizer & Company, 384 F.Supp. 265 (E.D.N.C.1974). While by no means intending to condone the activities of the defendants before the Patent Office, 41 the court is of opinion that plaintiffs have failed to carry their heavy burden of establishing fraud with respect to the ’119 and ’656 patents. Walker Process Equipment, Inc. v. Food Machinery and Chemical Corporation, 382 U.S. 172, 86 S.Ct. 347, 15 L.Ed.2d 247 (1965).

The principal defense advanced by defendants to plaintiffs’ patent misuse charges is that any misuse inherent in the original DMRC licensing agreement, which defendants continue to deny, was “purged” by the offer of new licenses after this litigation was instituted. 42 On December 23, 1970, two alternative license proposals were transmitted to the plaintiffs in which the licensee was given a choice between a license with a lump sum royalty for all licensed rights or a license with a variable rate which increased with each licensed right. On September 28, 1971, two additional license agreement proposals were transmitted to plaintiffs by DMRC (DX 687 through DX 703). The primary difference between the two sets of proposals was the abandonment in the second set of a requirement that the licensees contractually acknowledge the validity of the patents under which they were to be licensed. Each set of proposed licenses included unpatented “technical information” as well as patents.

In his letter of transmittal of the December, 1970 proposal Armitage of DMRC said:

“Enclosed are a copy of each of the two license forms for your information. It is *703 intended to have these licenses (for those who accept them) take effect as of November 1, 1969, which is before the date of the earliest litigation involving the validity and scope of the patents in suit.

“The new agreement will not, however, go into effect until 100% of royalties due prior to July 1,1970, under the particular license which is selected are paid in full, and 50% of royalties due from July 1, 1970, until the date upon which the new license is signed are also paid. For those parties who under the agreement selected have paid higher royalties than would be due under such agreement, a refund of the excess payment back to November 1, 1969, would be made.” (Armitage letter of December 23, 1970, DX 687.)

A similar condition was imposed in connection with Armitage’s letter transmitting the 1971 proposal. Aside from the fact that these proposed new license agreements carried forward the tying arrangement heretofore held to taint the package and by disproportionate royalty rates continued the coercive effects of the previous licensing scheme, the requirement that plaintiffs pay back royalties failed to dissipate the effects of the prior misuse and the proposed new licenses were therefore ineffective as a purge. 43 It is settled law that where patent misuse has been found the patentee is entitled to no accounting for past damages. Compton v. Metal Products Company, 453 F.2d 38, 46 (4th Cir. 1971).

Defendants’ second line of defense is grounded on the “single product” rule. Citing Times-Picayune Publishing Company v. United States, 345 U.S. 594, 73 S.Ct. 872, 97 L.Ed. 1277 (1953), and International Manufacturing Company v. Landon, Inc., 336 F.2d 723 (9th Cir. 1964), cert. denied sub nom., Jacuzzi Brothers, Inc. v. Landon, Inc., 379 U.S. 988, 85 S.Ct. 701, 13 L.Ed.2d 610 (1965), defendants argue that “no unlawful tie occurs if items are sold together as part of a single product.” The difficulty with defendants’ argument is that the ARCT machines themselves were not patented and the use of any one patent on the machine was not dependent upon the use of some other patented feature. Unlike the patents in International Manufacturing which were inherently interrelated so that the practice of one necessarily infringed another or necessarily required for its successful practice the practice of another, each of the Chavanoz apparatus patents could be practiced without the use of any other Chavanoz patent, and, of course, the practice of the process patents was not dependent upon the use of ARCT machines at all, much less the use of any particular Chavanoz apparatus patent incorporated in those machines. The defendants’ “single product” defense must be rejected.

Likewise unacceptable is defendants’ further contention that the plaintiffs were under an obligation to demand a license limited only to patents to be practiced and that failure to do so made the package licensing voluntary rather than mandatory and therefore not subject to the charge of patent misuse. 44 In the face of overwhelming evidence that demands by purchasers of the ARCT machines for something less than the license package offered would have been rejected by DMRC, how *704 ever, a specific demand by a plaintiff Throwster was not a prerequisite to a finding of a refusal to deal except on the basis of the tendered package. See United States v. Loews, Inc., 189 F.Supp. 373 (S.D.N.Y.1960), modified on other grounds, 371 U.S. 38, 83 S.Ct. 97, 9 L.Ed.2d 11 (1962).

“[T]he District Judge found that [plaintiff] never attempted to negotiate with ABC after it rejected its May 4th letter order. Judge Tenney seems to suggest that [plaintiff] had a duty to approach the hierarchy of officialdom at ABC’s New York office with respect to a discount on its selected lineup, and should not have been content to accept the rejection it received from the lower echelon in the network’s Chicago sales office. But the law does not demand that [a plaintiff] joust with windmills; it would have been reasonable to require it to negotiate only if there was a chance of success.” American Manufacturers Mutual Insurance Company v. American Broadcasting-Paramount Theatres, Inc., 2 Cir., 388 F.2d 272, 284-5 (1967).

After spending the better part of a summer “jousting” with the intransigent Armitage, Attorney McClure was finally able to obtain some minor concessions, but the patent package and level of royalty features of the DMRC license remained unchanged (PX 215 and 410). Similar efforts by other prospective licensees would doubtless have met with no success.

No discussion of the DMRC licensing program would be complete without some comment on its chief architect, Norman Armitage, and the impression that has been left with the court concerning his character and credibility. It is unfortunate that this most interesting and complex individual did not survive to testify in person at the trial, but from the mountainous volume of his deposition testimony and innumerable exhibits and the court’s own questioning of witnesses who knew him intimately the picture has emerged of a highly intelligent and sophisticated lawyer and executive who could be at once charming and convincing, domineering and duplicitous, pragmatic and pertinacious. 45 With undaunted zeal he pursued the DMRC license program as if it were the sole objective of his professional career, and the economic results achieved for his em *705 ployers attest the effectiveness of his efforts. Ironically, however, these same efforts led to charges of patent misuse and antitrust violations which spawned this Gargantuan litigation in which the court has now found the charges, at least in part, to be substantiated.

In summary, the court finds that while standing alone some phases of the conduct of the defendants do not support plaintiffs’ charges of patent misuse, the mandatory package licensing program and the totality of the conduct of defendants with respect thereto support the conclusion that the Chavanoz patents were misused, that the misuse has not been purged and that none of the Chavanoz patents, even if valid, is enforceable against the plaintiffs herein. 46 ,

*706 III.

THE PATENT VALIDITY AND INFRINGEMENT ISSUES

As previously indicated (Footnote 24) while the finding of antitrust liability might render unnecessary decision on the patent misuse, validity and infringement issues, the court has deemed it appropriate and in the interest of the efficient administration of justice to decide all of the issues in the case at this time. This appears to be the practice in other courts in similar cases. Union Carbide Corporation v. Borg-Warner Corporation, 430 F.Supp. 1, 2 (N.D.Ohio 1975), aff’d, 550 F.2d 355 (6th Cir. 1977). The court will therefore decide the patent validity and infringement issues with respect to each of the eight patents remaining in suit at the time of trial. 47 They are United States Patents Nos.:

2,891,375 3,165,881

2,944,319 3,232,037

3,012,397 3,283,414

3,123,973 3,584,450

A discussion of each of the patents follows.

Patent No. 2,891,875

Entitled “Apparatus for the Production of High-Bulk Yarn”, the ’375 patent was issued to Chavanoz as assignee of its two French employees, Louis VanDamme and Louis Rouyer, on June 23, 1959. 48 It is a combination patent covering the basic apparatus for processing yarn by use of the false twist method. The principal feature of the invention is a “curved tube heater” through which the moving yarn travels as it proceeds toward the false twist spindle and thence to the take-up package. 49 Pertinent portions of the specifications for this patent read as follows:

“This invention relates to apparatus for producing bulked yarn by imparting thereto and setting a false twist.

******

“The invention will be better understood by referring to the following description taken in connection with the accompanying drawing in which the figure is a diagrammatic view of one embodiment of the apparatus embodying the present invention.

“Referring to the drawing, the filament 1 from the winding 2 passes through a guide 3 and thence passes through a tube 5 which in the form shown is curved so that the filament contacts the surface of the tube as if is fed therethrough. In some instances, however, the tube 5 may be made straight instead of curved and the filament be caused to contact the side of the tube by suitable guide means. The tube 5 is provided with terminals 6 which are connected to a suitable source of electric current for heating the tube by its resistance effect.

******

“The filament after passing through the tube 5 is fed through a false twister 15 which is rotated by suitable means indicated as a belt 16 and is adapted to produce a twist in the filament which feeds back along the filament toward the heating tube 5 until it is arrested by contact of the filament with the surface of such tube. The temperature of the tube 5 is adapted to heat the filament for setting the false twist therein. Finally, the filament 1 passes through the guide 17 and is then wound onto a spool 18.”

*707 A copy of the drawing referred to in the specifications is shown below.

After two rejections by an examiner in the Patent Office three claims were finally allowed as follows:

“1. Apparatus for setting a false twist in synthetic thermo-plastic filaments comprising a heater tube composed of a peripheral wall completely surrounding a heating zone and made from metal having electrical resistance characteristics, said tube being curved axially into arcuate shape, means passing an electric current through said heater tube for heating said tube due to its resistance, means feeding said filaments through said tube in contact with the inner arcuate surface thereof, and a false twisting device disposed to impart a false twist to the filaments emerging from said tube, whereby the twist feeds along said filaments into the tube and is set therein and is removed after the filaments pass the false twisting device.

“2. Apparatus as set forth in claim 1 in which heat insulation is disposed around said tube.

“3. Apparatus for setting a false twist in synthetic thermoplastic filaments comprising a heater tube composed of a peripheral wall completely surrounding a heating zone, said tube being curved axially into arcuate shape, means heating the walls of said tube, means feeding the filaments through said tube in contact with the inner arcuate surface thereof, and a false twisting device disposed to impart a false twist to the filaments emerging from said tube whereby the twist feeds along said filaments into the tube and is set therein and is removed after the filaments pass the false twisting device.”

Plaintiffs contend that the ’375 patent is invalid for three reasons: (1) it is anticipated by Gilchrist Patent No. 2,958,921 (Gilchrist) and is therefore lacking novelty under 35 U.S.C. § 102(e); (2) it is the re-patenting of an old combination, and as such is void under the doctrine of Lincoln Engineering Company v. Stewart-Warner Corporation, 303 U.S. 545, 58 S.Ct. 662, 82 L.Ed. 1008 (1938); and (3) the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art, citing as such prior art in addition to Gilchrist Chavanoz’s own VanDamme Patent No. 2,761,272, Chavanoz’s Belgian Patent No. 539,620, the French Ancet Patent No. 1,090,011 and the Stoddard Patent No. 2,803,105. The plaintiffs also allege that the ’375 patent, even if valid, is not infringed by any ARCT machine, because on no machine is the twist *708 arrested in the heater as required by the patent.

Chavanoz and DMRC, on the other hand, vigorously assert that the ’375 patent constituted a major “breakthrough” in the field of false twist texturing; that far from lacking novelty, the curved tube heater represented a departure from the conventional wisdom which had taught that the moving yarn should not be allowed to contact the surfaces of the heater; that it filled a long-felt need for a heater that would produce uniform yarns which could be made into “barre-free” fabrics; that no prior false twist apparatus produced yarns of comparable quality or uniformity; and that machines embodying this invention met with unprecedented commercial success. On the question of infringement the defendants merely point to the fact that all ARCT machines sold in this country embody some version of the curved tube heater, and they deny that any significance should be attributed to the twist-arrest-within-the-heater feature mentioned in the patent.

Validity of the ’375 Patent

35 U.S.C. § 102(e) provides that:

“A person shall be entitled to a patent unless—

“(e) the invention was described in a patent granted on an application for patent by another filed in the United States before the invention thereof by the applicant for patent . . . ”

In an order entered herein on September 26, 1974, Judge Hemphill, while ruling that the Gilchrist patent could properly be considered on the question of anticipation, found a genuine issue of material fact precluding summary judgment, stating at the same time:

“It is important to keep in mind that the law requires the court to compare (1) the description of the Gilchrist, et al., patent with (2) the claims of the Chavanoz patent. One does not have to consider the claims of the Gilchrist, et ah, patent and one does not have to consider the disclosure of the Chavanoz patent. Rather one compares the claims of the Chavanoz patent with the disclosure of the Gilchrist, et ah, patent Whether or not the Gilchrist, et ah, ‘describes’ the closed curve tube heater claimed in the Chavanoz patent is a question for the trier of fact.”

The law appears settled that in order to anticipate an invention, “it is necessary that all the elements of the invention or their equivalent be found in one single description or structure where they do substantially the same work in substantially the same way.” 1 Deller’s Walker on Patents § 57, p. 242.

“Anticipation is strictly a technical defense. Unless all of the same elements are found in exactly the same situation and united in the same way to perform the identical function in a prior pleaded patent, there is no anticipation.” Stauffer v. Slenderella Systems of California, 254 F.2d 127, 128 (9th Cir. 1957).

With these principles in mind we begin our inquiry to determine whether every element claimed by the ’375 patent is disclosed by Gilchrist. If the elements are found to be identical, we must then determine whether they perform substantially the same function in substantially the same manner.

Like the ’375 patent, Gilchrist discloses an apparatus for producing textured synthetic yarns by the false twist method. It, too, is a combination patent employing the conventional devices of supply package, feed roll, heater, false twist spindle, output rolls and take-up package. Like the ’375, its principal feature is the heater. In the Gilchrist heater the yarn passes through a grooved, banana-shaped “tongue” the concave side of which is fastened at right angles to a pipe of about four inches in diameter through which a hot liquid is circulated. Thus heat is imparted to the tongue and the yarn running through it (PX 48). The grooved tongue is fitted with a flap or lid running longitudinally with it which lid is hinged at one end and shaped to the contour of the tongue. In the closed position this lid and the grooved tongue create a hollow passageway through which the yarn *709 travels. The yarn receives heat only from the tongue which is in contact with the pipe containing the circulating hot liquid.

The apparatus described above is shown in the following diagrams which have been reproduced from the Gilchrist patent.

In the side view of the apparatus illustrated in Figure 2 the yarn 13, having left the supply package (not shown), traverses a pulley 36 and enters the curved groove 10 which is attached to the pipe filled with hot liquid 11, proceeds thence through the false twist spindle 18 to a second pulley 35 and then onto a take-up package (not shown). In the cross section of the tongue shown in Figure 3 the Y-shaped groove 12 is surrounded by insulating material 15 and covered by the hinged flap or lid 14. (The small passage 16 leading from a water supply pipe 17 to the bottom of the curved grooved tongue shown in Figures 2 and 3 are not relevant for present purposes.) The entire apparatus showing electrical leads, etc., is shown in Figure 1.

It is not disputed that Gilchrist discloses all the elements of the broadest of the claims of the ’375 patent, claim 3, except those relating to the curved tube heater, and the question for decision, therefore, is whether Gilchrist discloses “a heater tube composed of a peripheral wall completely surrounding a heating zone, said tube being curved axially into arcuate shape, means heating the walls of said tube . . .” The court has concluded that it does.

Although the defendants strenuously point to the presence of uniform heat emanating from 360° around the yarn as constituting the novelty of the ’375 patent and the patent itself states that one of the principal advantages of the heater is “the fact that the convection radiation within the said tube, which appears to play an important part in the heating of the yarns continuously traveling through the tube, is sufficiently utilized,” claim 3 of the ’375 patent requires only a “means heating the walls of said tube.” That this function is satisfied by the Gilchrist apparatus which utilizes circulating hot oil for heating the yarn filaments passing through the tongue hardly seems open to question. 50 As Judge *710 Widener said in Technitrol, Inc. v. Control Data Corporation, 550 F.2d 992 (4th Cir. 1977):

“In the use of means clauses in patent claims, a ‘means for’ clause such as those used here ‘. . . in effect calls for structure — more precisely — for apparatus or, indeed, for any physical body or bodies having the capacity to perform the function recited after the v/ords ‘means for’. Such a clause is completely devoid of the details of apparatus capable of satisfying the recitation of function. Indeed, a ‘means clause’ (consisting of the words means for and a statement of the function which such means is supposed to perform) is to be construed as calling for ANY means capable of performing the indicated function.’ Rosenberg, Patent Law Fundamentals (1975), p. 48.” (Footnote omitted.) Id. at p. 999.

Nor is claim 3 rescued by the fact that the only heating means revealed in the specifications is by electrical resistance, for it has long been the law in this circuit that:

“While the claims should be construed in the light of the specification to obtain an understanding thereof, the limitations specified cannot be read into the claims.” Brown v. Brock, 240 F.2d 723, 728 (4th Cir. 1957).

To like effect see Graver Tank & Manufacturing Company v. Linde Air Products Company, 336 U.S. 271, 69 S.Ct. 535, 93 L.Ed. 672 (1949), and Wilcox Manufacturing Company v. Eastern Gas & Fuel Associates, 400 F.2d 960 (4th Cir. 1968). Defendants, having claimed broadly, may not now avoid Gilchrist by arguing that their patent only refers to an electrically heated tube that imparts heat evenly from 360° surrounding the yarn. 51

“[A patentee], having sought and obtained a broad construction of his claim, cannot [thereafter] narrow it so as to avoid anticipation by showing that the claimed method was used in a particular form of structure not claimed.” Smith v. Hall, 301 U.S. 216, 232, 57 S.Ct. 711, 718, 81 L.Ed. 1049 (1937).

Finally, under the doctrine of “claim differentiation” the presence of narrow claims, as are claims 1 and 2 of the ’375 patent, is evidence that limitations are not to be written into the broader claim 3. Smith v. Snow, 294 U.S. 1, 55 S.Ct. 279, 79 L.Ed. 721 (1934). This doctrine was most clearly stated in Texas Company v. Globe Oil & Refining Company, 112 F.Supp. 455 (N.D.Ill.1953), aff’d, 225 F.2d 725 (7th Cir. 1955), where it was said:

“The essence of the doctrine . is that a patentee is entitled both to narrow claims particularly directed to the preferred embodiment, and, to broad claims which define the invention without reference to the details of the disclosure; that in such case, the presence of the narrow claims negatives the limiting (by implication) of the broader claims; that the presence of an express limitation in one claim negatives an intent similarly to limit by implication a claim in which the limitation is not expressed; that the spelling out of the specific limitation in one or more claims gives special significance to the absence of such a specific limitation in another claim, because under such circumstances the attention of the applicant and the Patent Office is shown to have been directed to the limitation or to its absence, and the intent is clear that when the limitation was intended it was expressed.” Id. at p. 467.

*711 Thus for purposes of claim 3 of the ’375 patent the remaining question is whether the curved tongue with its hinged flap or lid as disclosed by Gilchrist is an element substantially identical to the “tube” as claimed in the ’375 patent. The Gilchrist patent discloses that “four heating units . are provided each of which comprises a curved tongue of metal such as copper or brass . . . each of the curved tongues is provided with a V-section groove extending along the length thereof and through which the yarn can be drawn;” that “the mouth of each V-groove is closed by a plate and the combination may be enclosed with suitable lagging;” and that in order to “prevent heat losses due to convection or radiation or due to the entry of air along the groove ... a flap may be hinged to the upper end of each tongue and shaped to the contour of the tongue and may project partly into the groove.” Other references indicate that the words “flap” and “plate” are used interchangeably, but whatever the hinged lid is called, it is clear that it may be shaped on the bottom side in a fashion so as to protrude into the V-groove and thereby narrow the passage through which the yarn passes.

Defendants assert that the triangular shape of the passageway through which the yarn passes in the Gilchrist heater precludes labeling it a tube. Plaintiffs’ expert testified, however, that it would be mechanically impossible to make a perfect V unless one uses two intersecting plates of metal, Tr. 5, pp. 918 and 932, and defendants’ expert agreed that mechanically speaking, the V-shaped groove disclosed by Gilchrist would be U-shaped, Tr. 60, p. 11,627. Although the court regards as overly-creative the suggestion by plaintiffs’ expert, Tr. 4, p. 718, that the underside of the hinged lid could be laterally curved so as literally to provide a circular channel through which the yarn could pass, the court is of opinion that the passageway afforded the yarn in the Gilchrist patent is “tubular” or “circular” within any commonly accepted definitions of those words. That the walls of the “tube” are not perfectly rounded does not, in the court’s opinion, create a material difference between the Gilchrist and ’375 embodiments. Claim 3 is anticipated by Gilchrist.

Claims 1 and 2 of the ’375 patent differ from claim 3 in that they require the tube to be made from metal having electrical resistance characteristics and specify that the means for heating the tube is by “passing an electric current through said heater tube for heating said tube due to its resistance.” Plaintiffs have acknowledged that Gilchrist does not disclose a method of heating by electrical resistance, Tr. 89, p. 17,133, and thus it would seem at first blush that Gilchrist clearly does not anticipate claims 1 and 2 of the ’375 patent. Plaintiffs contend, nevertheless, that heating a metal tube by passing electricity through it was well known in the art at the time the ’375 patent issued. Citing Butler v. Helms, 550 F.2d 954 (4th Cir. 1977), plaintiffs assert that “a merely extraneous structural feature recited in a claim, which has no function with respect to the invention claimed, should not prevent invalidity for anticipation if all of the requirements of anticipation are otherwise met as they are here.” Id. at pp. 956-7.

The court finds this argument unpersuasive. In Butler the validity of a patent for a sailboat canopy was in issue. The only structure claimed in the patent which was not disclosed in an earlier patent for a sailboat canopy was a different means of attaching the canopy to the mast of the boat. However, neither in the claims nor specifications was any virtue attributed to the different mode of attachment. Since this feature was non-inventive, even in the mind of the inventor, there was no difficulty in holding that its existence did not bar a finding of an otherwise clear anticipation. Such is not the case here. The specifications of the ’375 patent point to the uniformity of heat derived from the material of which the tube is made and the manner in which electricity is passed through it. Most of the defendants’ evidence on the inventiveness of this patent is addressed to this point. More importantly, there is no evidence that anyone had ever heated a *712 curved tubular element with electricity. Thus this is not the mere recitation of extraneous structure which led to the holding in Butler. Claims 1 and 2 of the ’375 patent are not anticipated by Gilchrist within the purview of 35 U.S.C. § 102.

The plaintiffs’ second ground of attack on the validity of the ’375 patent is that it is merely the re-patenting of an old combination with one element, the heater, perhaps improved. Plaintiffs’ position may be briefly summarized as follows: The ’375 patent is the sixth Chavanoz patent claiming a false twist apparatus combining a yarn feeding and guide means, a yarn heating means, a false twist means and a yarn take-up means. The only difference in the combinations is the specific heating means disclosed. Plaintiffs argue that because the heater in the ’375 patent performs no new function and produces no different result than any of the earlier heaters, it is obviously invalid as a re-patenting of an exhausted combination. Plaintiffs further argue that even if as a result of the improved heater in the ’375 patent a new result, that is, a better yarn, is achieved, the patents still cannot survive scrutiny under the doctrine in Lincoln Engineering Company v. Stewart-Warner Corporation, 303 U.S. 545, 58 S.Ct. 662, 82 L.Ed. 1008 (1938). Plaintiffs assert that under Lincoln Engineering improvement of one element of an old combination, even if it leads to the production of a better result by the entire combination, does not make the combination patentable unless the old elements of the combination perform their functions in new and different ways.

Defendants, on the other hand, argue that the doctrine of Lincoln Engineering is inapplicable to the ’375 patent for at least three reasons: “First, the elements of apparatus claimed in the patent cooperate with one another to achieve the object of the invention. Second, the patented apparatus achieves a new and improved result. Third, no one in our case has sought to enlarge their patent rights by suing sellers of old devices for contributory infringement of this patent.” (Appendix A to defendants’ post-trial brief.) Additionally, defendants assert that Lincoln Engineering is no longer sound law.

In Lincoln Engineering the patent involved was for an apparatus for lubricating bearings in which a nipple or fitting was connected with the bearing, a gun connected to a pump forcing lubrication under pressure, a hose connecting the pump with a fitting, and means of coupling the hose to the fitting to make a tight joint during greasing. The Supreme Court held this to be an old combination in the bearing lubrication art, and in holding the patent claim void the court said:

“As we said of Gullborg in the Rogers [Rogers v. Alemite Corporation, 298 U.S. 415, 56 S.Ct. 787, 80 L.Ed. 1251] case . . having hit upon this improvement he did not patent it as such but attempted to claim it in combination with other old elements which performed no new function in his claimed combination. The patent is therefore void as claiming more than the applicant invented. The mere aggregation of a number of old parts or elements which, in the aggregation, perform or produce no new or different function or operation than that theretofore performed or produced by them, is not patentable invention. And the improvement of one part of an old combination gives no right to claim that improvement in combination with other old parts which perform no new function in the combination.” 303 U.S. 545, 549, 58 S.Ct. 662, 664-65, 82 L.Ed. 1008.

Lincoln Engineering was applied by the Fourth Circuit in Heyl & Patterson, Inc. v. McDowell Company, 317 F.2d 719 (4th Cir. 1963). In that case the patent was for a combination titled an apparatus for loading bulk material into a ship’s hold. In declaring the patent void the court said:

“The patent is also invalid for claiming too much. The mere improvement of one portion of an apparatus does not permit a patenting of the whole [citing Lincoln Engineering and other cases]. Plaintiff, having at the very most hit upon a mechanical change to facilitate the use of *713 the upper end of the chute leading to the ship’s hold in place of a stationary pan, patented the entire apparatus from the hopper to the gate. This is much more than any possible invention to be found in plaintiff’s apparatus.

“We therefore, hold that the plaintiff’s patent is invalid in that it is lacking in invention and because it claims too much. The findings of the Patent Office and the District Court to the contrary are clearly erroneous.” Id. at p. 723.

Application of the rule in Lincoln Engineering to the facts of the present case has led the court to the conclusion that the ’375 patent is void. Like its five Chavanoz predecessors (Patents Nos. 2,741,893; 2,761,272; 2,780,047; 2,823,513; and 2,823,-514) the ’375 patent is simply a combination of the essential elements of the false twist process, all of them old in the art, the heater being the only element for which improvement was claimed.

Contrary to defendants’ argument, the heater element on the ’375 patent does not “cooperate” with the other elements within the meaning of the Lincoln Engineering doctrine, for cooperation with other elements means something more than the performance by each element of the same function in the same manner it was designed to perform. Normally all elements, old or new, must of necessity “co-act” or the machine will not operate. In order to support a finding of invention, however, it must appear that the co-action of the improved element with the old elements has a synergistic effect, that is, “[the] combination of elements [must] result in an effect greater than the sum of the several effects taken separately.” Anderson’s-Black Rock, Inc. v. Pavement Salvage Company, 396 U.S. 57, 61, 90 S.Ct. 305, 308, 24 L.Ed.2d 258 (1969). For instance, if a re-designed heater allowed the false twist spindle to operate at substantially greater speeds than had been possible with old heaters, the new heater and the old spindle would be “co-acting” in a manner different from old heaters and old spindles with the much desired and sought after result of higher production. No such showing has been made here. For all that appears from the evidence, installation of the ’375 heater on the false twist machines resulted in no change whatever in the function of the other elements on these machines. They continued to function in precisely the same manner as in earlier patented combinations.

It may be conceded, as defendants stress-fully contend, that machines with the ’375 heaters produced better quality and more uniform yarns. But here again this is not the test of invention. 52

In Electrolux Corporation v. Dust Pak, 215 F.Supp. 367 (E.D.N.Y.1963), aff’d, 330 F.2d 958 (2nd Cir. 1964), suit was brought for infringement of a patent for the construction of a disposable paper filter bag for use principally in vacuum cleaners of the horizontal cylinder type. The single *714 claim of the patent was upon a combination of certain of the structural elements that made up the bag and end member. Novelty was not claimed for any constituent element but for the union of the elements. In declaring the patent void the court said:

“Invention cannot be found in Anderson’s bringing to the familiar combination a better or, perhaps, a merely better defined bag element than was earlier in use. The ease is in principle within Bassick Co. v. R. M. Hollingshead Co., 298 U.S. 415, 425, 56 S.Ct. 787, 80 L.Ed. 1251 (1936); Lincoln Engineering Co. v. Stewart-Warner Corp., 303 U.S. 545, 547-548, 549-550, 58 S.Ct. 662, 82 L.Ed. 1008 (1938); Application of Lambert, 165 F.2d 441, 443, 35 Cust. & Pat.App. 839 (CCPA 1948). There is not room to argue that Anderson brought into play any new principle or change in the method of co-action of the elements of the old combination. He brought to it a better bag as a bag and it was better wholly within the functions and attributes that belonged to it as a bag. The improvement in the combination was solely derived from the increment in an unchanged kind of utility within that one element of the familiar combination and not in anything peculiar or specific to the union of elements that is the genius of the claim. Concededly, or if not so, then manifestly, there is not here anything distinctly patentable as an improvement in the carrying-bag art, or indeed in any distinct element of the combination. What has been brought to the combination is not an infusion of fresh invention — but good manufacturing practice and a good choice of a bag from the obviously pertinent carrying-bag art.” Id. at p. 370.

In order to be patentable in the first place a machine or an improvement thereof must be “new and useful”, 35 U.S.C. § 101, but the mere factors of novelty and utility in one element of a machine do not justify a repatenting of the entire machine consisting of a combination of the new element with old elements whose functions remain unchanged.

Two Fourth Circuit cases cited by defendants for the proposition that a new result by itself is sufficient to escape the prohibition against re-patenting old combinations, Marvel Specialty Company v. Bell Hoisery Mills, 330 F.2d 164 (4th Cir. 1964), and Colgate-Palmolive Company v. Carter Products, 230 F.2d 855 (4th Cir. 1956), are not controlling here. These cases simply state the settled rule that even though all of the elements in a combination patent may be known to the prior art, the combination itself may be patentable if it achieves a non-obvious result. Neither case deals with the situation that arises when such a combination has earlier been patented and a later patent attempts to re-patent the entire combination based on the improvement of only one element.

Two other cases, Porter-Cable Machine Company v. Black & Decker Manufacturing Company, 274 F.Supp. 905 (D.Md.1967), and Reese v. Elkhart Welding & Boilerworks, Inc., 447 F.2d 517 (7th Cir. 1971), cited by defendants in support of the proposition that a new result alone is sufficient to justify the re-patenting of an old combination do not squarely support defendants’ position when read closely, for in both cases it was expressly found that the old elements interacted with the improved element in a new fashion to produce the new result.

In a third argument advanced by defendants they undertake to distinguish Lincoln Engineering on the grounds that it involved a suit for contributory infringement. In their post-trial brief they say:

“Our suit, on the other hand, is not for contributory infringement. We assert infringement of U. S. Patent 2,891,375 only against Throwsters who use in combination all of the elements of apparatus claimed in the patent. Neither DMRC nor Chavanoz has ever asserted that apparatus comprising less than all of the elements claimed infringes or contributorily infringes this patent. On the contrary, we readily admit that no one infringes this patent by the use of any apparatus or combination that does not *715 include the novel improvement claimed in the patent. Therefore, Lincoln Engineering and Bassick are inapplicable.”

For this proposition defendants cite Williams Manufacturing Company v. United Shoe Machinery Corporation, 316 U.S. 364, 62 S.Ct. 1179, 86 L.Ed. 1537 (1942). In that case the Supreme Court affirmed a judgment upholding the validity of a patent for improvements in the automatic heel lasting mechanism in a shoe making machine. The alleged infringer, relying on Bassick Manufacturing Company v. R. M. Hollingshead Company, 298 U.S. 415, 56 S.Ct. 787, 80 L.Ed. 1251 (1936), and Lincoln Engineering, contended that the patentee could not “by improving elements of an old combination whose construction and operation is otherwise unchanged except as modified or improved by the adjustments, in effect re-patent the old combination for another seventeen years by reclaiming it with the improved elements substituted for the old elements,” and that the courts below had erred in holding the contrary. Responding to this contention the Supreme Court said:

“The contention is not in accord with the holdings below ... In respect of each claimed combination, both courts have found that they embody other improvements, in addition to mere manual preliminary adjustments, and that each combination exhibits invention in that its elements cooperate in a new and useful way to accomplish an improved result.

“[E]ach of the claims is confined to a combination of specified means applicable only to a restricted portion and function of the whole machine. In stating his claims, the patentee sometimes says ‘a machine of the class described having, in combination, . . . ’ Obviously, no machine will infringe which does not have in combination the means specified in each of the claims for accomplishing the particular portion of the total operation covered by the claim. .

“Such preliminary statement is commonly and properly used to specify the type of machine in which the claimed subsidiary combination of elements works an improvement over the prior art. In describing the novel combinations embodied in the claims, it was necessary to make reference to certain portions of the machine in connection with which the new combinations were to operate and with which they were to dovetail, but, in mentioning these other mechanical parts, the claim does not purport to embody them as elements of the claimed combination.” 316 U.S. 364, 368-9, 62 S.Ct. 1179, 1181-82, 86 L.Ed. 1537.

It is true that in holding Bassick and Lincoln Engineering not applicable to the facts in Williams Manufacturing the court said that they were suits for contributory infringement. But it is clear from the context in which the statement was made that this was not the grounds of the distinction. Rather, it was the fact that in Bassick and Lincoln Engineering the patentees, unlike the patentee in Williams Manufacturing, had undertaken to repatent all of the elements of the combination without regard to whether the improved element cooperated with the old elements “in a new and useful way to accomplish an improved result.”

“The effort was to extend the monopoly embodied in the improved pin fitting so as to prevent sale or use of well known grease guns of the prior art, although whatever was novel in the improved pin fitting was peculiar to itself and had nothing to do with the grease gun commonly used in connection with all pin fittings. Had the claim merely recited that it applied to an improved type of pin fitting to be used in connection with grease guns and couplers, the claim for the pin fitting would have been good, and would not have been rendered bad by the statement that the fitting was intended for use in connection with those other instrumentalities.” 316 U.S. 364 at p. 370, 62 S.Ct. 1179 at p. 1182, 86 L.Ed. 1537.

And so it is here. Had the claims of the ’375 patent simply recited that it applied to an improved type of heater to be used in connection with the other elements *716 of a false twist machine (feed rolls, false twist spindles, etc.), the claim for the heater (assuming it otherwise satisfied the statutory requirements for invention) would not have been rendered bad by the statement that it was intended for use with these other instrumentalities.

Defendants’ attempt to distinguish Lincoln Engineering on the grounds that it involved a suit for contributory infringement is rejected.

Defendants’ final argument on this issue is that Lincoln Engineering is no longer good law. They find support for this assertion in In re Bernhart, 417 F.2d 1395, 57 CCPA 737 (1969), and in a more recent decision by the Patent Office Board of Appeals, Ex Parte Barber, 187 U.S.P.Q. 244 (1974).

In Bernhart the examiner had rejected a combination claim under 35 U.S.C. § 112. 53 After quoting Section 706.03(j) of the Manual of Patent Examining Procedure to the effect that an applicant who has improved one element of a combination “which may be per se patentable” is not entitled to a claim to the improved element in combination with the old elements where the elements perform no new function in the combination and conceding that this rule has the support of many cases including Lincoln Engineering, the court went on to say:

“Why an applicant should not be so entitled is not clear. Many cases have said that the combination claim reciting only one new element with no new result is overclaiming or claiming more than the applicant invented. Such statements are indeed puzzling in view of the fact that the addition of elements to a claim nar rows its scope and thereby creates a lesser monopoly. Others have said the combination is not new, or is obvious, if no new coaction or result is obtained. This too is unsound, since it is not the result which is to be patented but the recited machine, composition, etc. If the prior art does not show or suggest the improved element itself, it defies logical reasoning to say that the same prior art suggests the use of that improved element in a combination.” 417 F.2d 1395 at p. 1402.

On the basis of this reasoning the court declined to follow Lincoln Engineering and concluded that the rejection of the claims in question on the ground of old combination under 35 U.S.C. § 112 was erroneous.

This decision was followed in Ex Parte Barber, supra, a two-one decision of the Patent and Trademark Office Board of Appeals. In each case it was held that the provisions of Section 112 had been satisfied, 54 and the Board of Appeals in Barber went on to express its agreement with the Court of Customs and Patent Appeals in Bernhart in declining to follow Lincoln Engineering, stating:

“Bearing in mind however that Lincoln Engineering came to the court on an issue of contributory infringement, we think it fair to assume that the court may have had in mind the possibility of a party somehow being held liable for making, using or selling a nonpatented article merely because it ultimately became part of a patented combination. Such fears, whether or not warranted in 1938, clearly have no basis under today’s statute, since *717 35 U.S.C. 271(c) limits contributory infringement to ‘a material part of the invention’ which is ‘especially made or especially adapted for use in an infringement of such patent, and not a staple article or commodity of commerce suitable for substantial non-infringing use’. Thus, we think it may reasonably be said that by incorporating this statute into the Patent Act of 1952, a statute having no prior counterpart, Congress removed the underlying rationale of Lincoln Engineering and thereby effectively legislated that decision out of existence.” Ex Parte Barber, 187 U.S.P.Q. 244, 246 (1974).'

In a well reasoned dissenting opinion Examiner in Chief Magil took issue with this hypothesis, pointing out that notwithstanding the amendments to the patent law in 1952, the courts, with the exception of the Court of Customs and Patent Appeals in Bernhart, have continued to follow Lincoln Engineering and that no court has cited Bernhart for the proposition that Lincoln Engineering is no longer viable. Concluding that the question should be clarified at a higher level, Examiner in Chief Magil said:

“If the Bernhart decision were the last word on this subject, I would have no hesitation in reversing the rejection in the present case because I firmly believe that a proper judicial system requires that a lower tribunal follow the principles laid down by a higher tribunal. However, the history of ‘old combination’ during approximately the past five years cannot be disregarded.” 187 U.S.P.Q. at p. 247.

This court agrees. In the almost eight years which have elapsed since Bernhart was decided no other court has followed it on this point nor has any other court concluded that Lincoln Engineering does not remain good law. It was followed in the Fourth Circuit long after the patent laws were rewritten in 1952 in Heyl & Patterson, Inc. v. McDowell Company, 317 F.2d 719 (4th Cir. 1963), a ease which, incidentally, did not involve contributory infringement. Notwithstanding the seeming harshness of the rule in Lincoln Engineering, the policy of the patent law which “visits total forfeiture upon a patent which attempts to extend its monopoly to something already patented” remains unchanged. Holstensson v. V-M Corporation, 325 F.2d 109, 122 (6th Cir. 1963).

Indeed, the fact that the policy underlying the rule in Lincoln Engineering was first stated by the Supreme Court of the United States one hundred fifty years ago in Evans v. Eaton, 7 Wheat. 356, 20 U.S. 356, 5 L.Ed. 472 (1822), affords this court ample reason for declining the suggestion that it is no longer good law absent statutory change or a pronouncement by higher authority binding on this court. 55

“In contemplating this seemingly drastic rule, we bear in mind that there is no natural or common law right to a monopoly of an invention. . . . Applicants for such monopolies must comply with the requirements that Congress insists be met as a condition to exercising such monopoly. The seeming harshness of this rule is further ameliorated when we keep in mind that an inventor may gain monopoly for an inventive improvement to an existing device simply by limiting his claim to that which he did invent.” Holstensson, supra, at 125.

In summary, the ’375 patent represents nothing more than the sixth repatent *718 ing by Chavanoz of an old combination with one element, the heater, improved with the other elements performing the same functions in identical fashion in each of the “inventions” and with any new result in the operation of the overall apparatus being directly attributable to the improvement in the heater element. The patent is condemned under the doctrine in Lincoln Engineering.

The plaintiffs’ third ground of attack on the ’375 patent is that it is void for obviousness under 35 U.S.C. § 103 which reads:

“A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.”

The court has concluded that the patent is void for obviousness under this statute.

The ’375 patent on its face discloses initially that it is the fourth in a series of patents by the same inventors on a false twist texturing apparatus. According to the patent, “an object of the present invention is to provide a further improvement in which the heating element is a mental tube of small diameter which constitutes an electric resistor.” The patent goes on to say:

“The principal advantage of the resistor-type heating tube is that it is easier to produce with uniform resistance properties than the U-blade [also known as the “umbrella rib”], which is a tubular segment and the fact that the convection radiation within the said tube, which appears to play an important part in the heating of the yarns continuously traveling through the tube, is efficiently utilized.

“For all these reasons, the output of the electric heating is clearly improved and the heat losses are reduced to a minimum.”

Three claims were finally allowed by the Patent Office as set forth in full on pages 707, supra.

As discussed in the previous section on validity, the principal features of claims 1 and 2 are “a heater tube composed of a peripheral wall completely surrounding a heating zone and made from metal having electrical resistance characteristics, said tube being curved axially into arcuate shape, means passing an electric current through said heater tube for heating said tube due to its resistance,” and claim 3 differs only in that the means for heating the tube is not limited to electrical resistance. Our task, then, is to determine if the differences between the subject matter sought to be patented and the prior art were such that the subject matter as a whole would have been obvious to a person having ordinary skill in the art at the time of the invention.

The oft-repeated test for obviousness under 35 U.S.C. § 103 is that

“[T]he scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined.” Graham v. John Deere Company, 383 U.S. 1, 17, 86 S.Ct. 684, 694, 15 L.Ed.2d 545 (1966).

Scope and Content of the Prior Art.

Although the plaintiffs cited a large number of patents against the ’375 patent as establishing the scope and content of the prior art, in their proposed findings of fact they rely on only five patents as prior art: 56 *719 United States Patents Nos. 2,803,105 (“Stoddard ’105”, PX 32); 2,958,921 (“Gilchrist ’921”, PX 47), and 2,761,272 (“Vandamme ’272”, PX 4); French Patent No. 1,090,011 (“Ancet ’Oil”, PX 42-A), and Belgian Patent No. 539,620 (“Belgian ’620”, PX 44). Of these the Vandamme ’272, Stoddard ’105 and Ancet ’011, along with the United States Patents 2,813,393 (Kings-burg ’393) and 2,823,292 (Kunzle ’292) were cited to the Patent Office. Gilchrist ’921 and Belgian ’620, the latter well known to Chavanoz, were not cited.

The Stoddard, Ancet and Kunzle patents establish that at the time of the invention described in the ’375 patent straight false twist machine heater tubes of small diameters were well known in the art. Curved contact heaters were equally well known, having been disclosed in the Gilchrist, Ancet, Vandamme and Belgian ’620 patents as well as in the Lewis patent (United States Patent No. 2,199,411, PX 18). Two of these patents, Gilchrist and Vandamme, disclosed that the curved contact heater should be covered in order to prevent heat loss and temperature variation due to ambient air conditions, and the Belgian ’620 patent disclosed that the curved heating strip should be surrounded by a cylinder of insulating material. Electrical resistance as the source of heat supply to a false twist machine heater was also well known, having been disclosed in the Lewis, Ancet, Belgian ’620 patents and one of DMRC’s own patents, Klein Patent No. 2,777,276 (PX 31).

As a starting point, therefore, it should be borne in mind that at the time of the application for the ’375 patent, contact heating, straight tube heating, and heating by electrical resistance were each individually disclosed by the prior art. We turn, then, to a more detailed consideration of the five patents actually relied on by plaintiffs as invalidating the ’375 patent for obviousness.

Stoddard ’105. This patent discloses an apparatus for processing thermoplastic textile yarns. It is not primarily a false twist machine, but rather an “uptwister” in which the yarn is taken off the machine in a twisted condition and later run through a machine again to untwist it before it is commercially usable. Of particular interest is the heater which is described as “a tube of non-ferrous material and of small diameter and bore, through which the yarn is passed, said tube being exteriorly insulated with glass or other electrical insulation and surrounded by coiled resistance wire.” Later on the patent discloses that “it is to be particularly noted that the winding pitch of the heating coil increases progressively from the bottom of the tube at which the yarn enters to the mid height of the tube and progressively decreases at a similar rate toward the top or exit end of said tube. As a result, a greater amount of electrical energy is provided at the entrance and exit ends of the tube making possible the maintenance of a desired uniform temperature throughout the tube.” The patent further discloses that the apparatus should be equipped with voltage regulators and thermostatic sensing means “of any approved standard and commercially available types by cooperation of which the temperature and the heating device is modulated compensatively with changes in ambient or room temperature and transfer of heat to a yarn traveling therethrough.” The commercial embodiment of this patent was the Fluflon machine marketed by Leesona.

Ancet ’Oil. This patent was sometimes referred to in the trial as “The Antique Shop” because of its comprehensive claiming of all that had gone before it in the field of uptwisting and false twisting. Again, the heating unit is of particular interest. The patent discloses that yarn in the false twisting process may be heated in “any suitable manner”; however, it is “preferably performed by causing the yarn to rub against a heating plate, for instance, a plate of polished metal heated by an electric resistor.” The patent further discloses that the heating device can be similar to that claimed by Stoddard, for one may “have the yarn traverse a tube electrically heated to such a temperature that the walls *720 heat up said yarn by radiation to the desired temperature. . . . Such a device makes it possible to achieve a very even heating of the yarn through the strict compliance with a certain number of conditions, that is to say, a strict control of the temperature of the wall of the tube, and absolutely constant speed of the yarn and the absence of eddy air currents capable of disturbing the operation.” The diagrams in this patent disclose both variations, both a straight tube radiant heater and an open-faced contact heater.

Ancet ’Oil as well as Stoddard ’105 were cited to the patent examiner, and the claims of the ’375 patent were initially rejected based largely on Ancet. The file wrapper shows that originally Chavanoz sought to patent a straight tube contact heater, with a dependent claim 3 in which the heater was curved. Claim 1 (and claim 2, which throughout has only involved the addition of insulation around the heater tube) was rejected as obvious in the light of Ancet’s disclosure of both an electrical resistance contact heater and a tubular heater. Claim 3 was rejected as being unpatentable over Ancet, especially in light of the fact that Chavanoz’s own Vandamme ’272 patent, to be discussed later, disclosed a curved heater.

Gilchrist ’921. This patent has been discussed in some detail in the previous section on validity. Recapitulating, the heater apparatus disclosed therein is a curved tongue of metal, with a V-shaped groove slightly rounded at the bottom through which the yarn travels. A hinged lid or flap fits over the top of the tongue, and it may be shaped so as to protrude down into the groove and further narrow the passageway for the yarn. The system is heated by means of hot oil passing through a metal tube to which the tongues are attached. According to the patent, the reason for enclosing the yarn is to “prevent heat losses due to convection or radiation or due to the entry of air along the groove.” As previously stated, Gilchrist does not anticipate claims 1 and 2 of the ’375 patent only because that patent claims a specific heating means (electrical resistance) not embodied in Gilchrist.

When analysis is made of the final argument made to the patent examiner before the ’375 patent issued it is seen that Gilchrist is the most pertinent prior art in this area. Following the rejection of the first application based largely on the Ancet ’Oil patent, Chavanoz refiled basically the same application with some cosmetic changes. The application was again rejected, this time as being unpatentable over Kingsburg, a patent never mentioned thereafter, and Kunzle, who disclosed a straight electrically heated tube. Again claim 3, which at this point was still the curved embodiment of the straight tube contact heater disclosed in claim 1 of the application, was rejected as simply being unpatentable in view of the Vandamme ’272 patent that disclosed a curved contact heater. In the last arguments made to the patent examiner, the solicitor distinguished the Vandamme curved contact heater as rapidly dissipating the heat, presumably because it is open. It seems highly likely that if the examiner had had the Gilchrist patent before him at that time, the fact that it is a closed curved heater would have affected his conclusion.

The Vandamme ’272. The heating means on this particular false twist apparatus patented by Chavanoz is a hollow curved metal casing with hot oil passing through the inside of the casing and the yarn being heated by passing over the outside. According to this patent the curvature of the metal casing is such that the yarn is “caused to contact the casing for a substantial distance in its travel.” In addition, the patent states that “excessive heat loss is prevented by a reflecting insulated cover plate which is disposed on the side of the casing over which the yarn passes and reflects heat from the casing onto the yarn.” As mentioned earlier, the patent examiner rejected the ’375 claims twice on the grounds that no invention was involved in combining the straight tube heaters known through various pieces of prior art and the curved contact heater disclosed in the ’272 patent.

The Belgian ’620. This is the Belgian counterpart of United States Patent 2,780,- *721 047 (the ’047 patent), which disclosed the “umbrella rib” heater. Although the co-pending application for the ’047 patent was disclosed in the application for the ’375 patent, the fact that the ’620 patent had issued in Belgium in time to be considered as prior art against the ’375 patent was not disclosed to the examiner. Defendants, citing the rule against double patenting, argue that this was harmless error, because the examiner would not have issued both patents if either was obvious in light of the other.

The cases cited by defendants on this issue indicate that double patenting is a complicated and convoluted doctrine. Contrary to plaintiffs’ assertion, the Court of Customs and Patent Appeals has stated that it is “a firmly established rule in this court for over thirty years that when the applicant takes out a patent on one of two or more co-pending applications on closely related inventions, he shall not be allowed claims in the other applications except on subject matter which is patentably different from the subject matter claimed in his earlier patent.” Application of Zickendraht, 319 F.2d 225, 50 CCPA 1529 (1963). Therefore, it appears that once the ’047 patent issued a patent examiner would not have been free thereafter to issue the ’375 patent unless he believed it was non-obvious over the ’047 patent. In addition, a patent examiner may rely on the claims of a patent issued on a co-pending application, in combination with other pieces of prior art, to reject a later application from the same inventor. Application of Simmons, 312 F.2d 821, 50 CCPA 990 (1963).

Given these rules, however, there is one basic difference in the effect that the examiner was free to give to the co-pending application for the ’047 patent as contrasted with the effect he would have been able to give to the Belgian ’620 patent had it been cited to him. The rule is that a patent resulting from a co-pending application by the same inventor may be relied on to invalidate a later application only insofar as its specific claims render the later application obvious. This differs from a consideration given to a piece of prior art, for in such case all that is disclosed in the prior art may be considered in making the obviousness determination. Application of Simmons, supra, 312 F.2d at 823. Because of this difference and because of the general unsettled nature of the law of double patenting, defendants should not be able to rely on it to excuse their failure to cite their own Belgian ’620 patent to the examiner as pertinent prior art.

Turning to the substance of the Belgian patent, we find therein disclosed an apparatus for false twist texturing in which the heater consists of a curved strip of metal that is U-shaped in cross section, with the yarn being heated by contact with the bottom of the U. The heater is made of electrically resistant metal, and the means of heating is by passing electricity directly through the heating element. The patent further teaches that in order “to prevent heat losses, the shaft or strip (which is the heating element) may be contained in a cylinder made of insulating material . and provided with a longitudinal slot opposite the hollow part of the strip or shaft.”

In summary, these five patents support the conclusion that at the time of the application for the ’375 patent contact heating elements, tubular heating elements, and insulated electrical resistant heating elements were all well known to the art. 57

*722 Differences Between Prior Art and the ’375 Patent.

The defendants have summarized their contentions as to the structural differences between the prior art and the ’375 patent as follows:

1. “It maximized the efficiency of heat transfer to the yarn by using conduction, convection and radiation heating in a way that was superior to any other heater design.”

2. “Conduction heating was achieved by the curvature of the tube which forced the yarn into almost continuous contact with the heater walls. (In grooved heaters, such as Scragg and Leesona, the yarn could and did jump out of the track to a place where it would receive less or no contact heat; if it jumped in an ARCT heater, it could land only on a heat generating surface.)”

3. “Convection heating resulted from the contact of the yarn with hot gasses swirling in the tube — gasses which retained their heat because they were surrounded or trapped by heater walls which were constantly generating heat;” and

4. “Finally, radiation heating was maximized again by the fact that heat was generated all around the yarn.

“In short, the curved tube design of the ’375 patent combined in one heater every advantage of all other heaters known to the art.” Defendants’ proposed findings of fact III — 58.

The question thus becomes, could a person of ordinary skill in the art, presumed to be familiar not only with the heaters already in commercial existence but also with the teachings of Chavanoz’s own Vandamme ’272 and Belgian ’620 patents that were not embodied on any machines at that time, have hit upon the idea of combining all of the advantages of these various heaters into the curved tube heater? The testimony of defendants’ expert discloses that in fact the advantage claimed for the ’375 heater by defendants is the manner in which it combined well known features of the prior art to achieve the most efficient result possible. Tr. 59, 11580-82.

For reasons not quite clear to the court defendants throughout have limited their consideration of the relevance of the prior art to heaters that were actually embodied on commercial machines at the time the ’375 patent was applied for. 58 They simply ignore for the most part the existence of earlier patents in this area and the pertinent teachings found within them. Thus although their conclusion about the conventional wisdom being employed by Leesona and other machinery manufacturers may be correct, it ignores the fact that Chavanoz in the Belgian ’620 patent of a year earlier had already abandoned the conventional wisdom by teaching that the yarn should pass through a curved heater surrounded by an insulating cylinder. Also, the Gilchrist patent, with a priority date at least two years earlier, had abandoned the conventional wisdom insofar as it taught that the yarn should be fully enclosed while the machine was in operation. The patent-ability of the ’375 patent cannot be sustained on the ground that it was the first commercially satisfactory heater.

In distinguishing the ’375 patent from the prior art defendants’ strongest argument, and the argument on which this patent must rise or fall, is that its structural features achieved the non-obvious result of allowing the heater temperature to be closely regulated within one position and from position to position across the ma *723 chine. Defendants’ position is that even if it might have been obvious to a man of ordinary skill in the art to shape a heater like that disclosed in the ’375 patent, he would have refrained from doing so because nothing in the prior art taught him that such a heater would have any particular advantages over any of the earlier models. Basically, all of defendants’ proposed findings of fact go to this issue. They create a picture of an industry desperately searching for a heater that could provide a uniform, quality yarn and they assert that the ARCT machine embodying the ’375 heater obtained quick and universal recognition as the solution to this problem.

Plaintiffs’ on the other hand assert that the ARCT machine produced yarn comparable to that produced on its competitors’ machines and that its quality was at any rate attributable to features other than the heater. Therefore, in their view, the ’375 heater is simply an obvious rearrangement of prior art metho.ds producing no different result.

There are reams of testimony on this issue of extraordinary, nonobvious results, much of it conflicting, but it would unduly prolong this memorandum to analyze in detail this voluminous testimony. The court must therefore content itself with the summary findings that the Throwsters had much the same problems with heat uniformity using the ’375 heaters as they did with other commercial heaters, and the extent to which the ARCT machines employing the ’375 heater were able to produce uniform yarns of high quality was due largely to the sophisticated electrical controls in the later machines and to the manner in which the heaters on the earlier machines were electrically calibrated. It is clear that defendants cannot rely on any uniformity attributable to the sophisticated electrical controls since these were in fact manufactured and patented by another company, Telemechanique, and since the control mechanism is not disclosed or claimed by the ’375 patent.

The conclusion is that this patent covers simply an alteration of existing structural components and that in and of itself it is responsible for no non-obvious result.

Level of Skill in the Art.

Defendants seek to impart vitality to the ’375 patent by pointing to the low level of skill in the art at the time of the patent application. It is undoubtedly true that very little sophisticated research was going on during this time, with advances being made on the assembly lines by men who were not highly educated in a formal sense. Thus it may be that if any one of these men was singled out and quizzed as to whether the ’375 design would have been obvious to him, he would have replied that it was not. However, it is important to keep in mind that the statutory test does not relate to any particular inventor working at a particular time, but rather relates to a hypothetical inventor presumed to have knowledge of all that has gone before. As Judge Widener said in Blohm & Voss v. Prudential-Grace Lines, Inc., 489 F.2d 231 (4th Cir. 1973):

“The inventor who ventures into the art is presumed to know the prior art references. Judge Rich stated it picturesquely when he wrote that the inventor should be imagined as seated in his workshop ‘with the prior art references— which he is presumed to know — hanging on the walls around him . . . .’ Application of Winslow, 53 CCPA 1574, 365 F.2d 1017, 1020 (1966).” Id. at p. 244.

And by the same token in a case such as this where old elements have been combined to produce an improved result, can it not be said that there should be attributed to the hypothetical inventor the combined skills of the inventors of the several components?

In this case, because the defendants are accurate in their portrayal of how research in this area was being carried out, it is highly unlikely that any particular person skilled in the art during this time period had any means of keeping abreast of rapid developments occurring in several countries and reflected in patents issued in several *724 languages. 59 The statutory test that imputes full knowledge to a person skilled in the art before determining obviousness cuts harshly against arts that are developing informally in remote places, for there is more of a discrepancy between any particular inventor and the statutory model. In this case the fact that the incorporation of previously well known elements into a curbed tube heater did not occur to any particular person prior to Chavanoz’s inventors and the further fact that several persons, including one of plaintiffs’ own witnesses, testified that it would not have been obvious to them at that particular time, does not establish non-obviousness to the statutory man of skill in the art.

The court has given consideration to defendants’ arguments based on the “secondary” considerations of long-felt need, immediate copying and commercial success, but as said by the Second Circuit in the recent case of Digitronics Corporation v. New York Racing Association, Inc., 553 F.2d 740 (1977), at page 748,

“Despite the persuasive advocacy of Judge Learned Hand, Reiner v. I. Leon Company, 285 F.2d 501 (2nd Cir. 1960); Lyon v. Bausch & Lomb Optical Company, 224 F.2d 530 (2nd Cir. 1955), this view does not represent current law. Any theory that ‘secondary’ considerations must be given weight before a determination of obviousness can be made was laid to rest in Sakraida v. Ag Pro, Inc., 425 U.S. 273, 282-83, 96 S.Ct. 1532, 47 L.Ed.2d 784 (1976), where Mr. Justice Brennan concluded for a unanimous Court:

‘Though doubtless a matter of great convenience, producing a desired result in a cheaper and faster way, and enjoying commercial success, [the invention] “did not produce ‘a new or different function’. . . . within the test of validity of combination patents.” Anderson’s-Black Rock v. Pavement Company, supra, [396 U.S.] at 60, [90 S.Ct. 305]. These desirable benefits “without invention will not make patentability.” Great A & P Tea Company v. Supermarket Corporation, 340 U.S. [147] at 153, [71 S.Ct. 127, 95 L.Ed. 162].’

“ . . . Only in a close case, in which application of the subjective criteria of nonobviousness in 35 U.S.C. § 103 does not produce a firm conclusion, can these objective or secondary considerations be used to ‘ “tip the scales in favor of patentability.” ’ ” Id. at p. 748.

And, of course, the court has considered the statutory rule that patents regularly issued enjoy a prima facie presumption of validity, 35 U.S.C. § 282. But this presumption is weakened if not indeed destroyed when it is shown that there existed prior art which the patent office did not consider. Blumcraft v. Citizens & Southern National Bank, 407 F.2d 557 (4th Cir. 1969); Heyl & Patterson v. McDowell, 317 F.2d 719 (4th Cir. 1963). Plaintiffs here have shown that at least two pieces of prior art, Gilchrist and Belgian ’620, more pertinent to the subject invention than that considered was not called to the attention of the Patent Office. Kalkowski v. Ronco, Inc., 424 F.Supp. 343, 352 (N.D.Ill.1976).

In summary, the court is of opinion that in the light of the five patents cited by plaintiffs as prior art, a heater for a false twist machine built according to the teachings of the ’375 patent would have been obvious at the time of the invention to a person having ordinary skill in the art. Accordingly, these teachings were unpatentable under 35 U.S.C. § 103 and the patent is invalid for obviousness.

Infringement of the ’375 Patent.

If the court is correct in its conclusion that the ’375 patent is invalid, the question of infringement becomes moot, for an invalid patent cannot be infringed. Bendix Corporation v. Balax, Inc., 421 F.2d 809 (7th *725 Cir. 1970). For reasons previously stated, however, all issues in the ease are being considered and decided. 60

Defendants as counterelaimants in their suits for damages for infringement have the burden of proof on the issue. They allege quite simply that every ARCT false twist machine sold in this country since 1959 has been equipped with a curved tube heater based on the ’375 patent; that all of the plaintiff Throwsters own and operate these machines; and that infringement is therefore established as a matter of law.

Plaintiffs, on the other hand, argue with equal assurance that their ARCT machines do not infringe the ’375 patent. This is so, they assert, because the patent, when read as a whole in conjunction with the record of the patent’s prosecution as reflected in the file wrapper, claims for the heater a function vital to the operation of all false twist machines, that is, the function of arresting at some point between the heater and the yarn supply package the twist in the yarn as it travels backward from the false twist spindle. Plaintiffs maintain that having claimed the entire false twist machine combination and having disclosed no separate means for arresting the twist, the ’375 patent claims this function for the heater itself as evidenced by this language from the specifications:

“The filament after passing through the tube 5 is fed through a false twister 15 which is rotated by suitable means indicated as a belt 16 and is adapted to produce a twist in the filament which feeds back along the filament toward the heating tube 5 until it is arrested by contact of the filament with the surface of such tube.” (Column 2, Lines 9-14 of the patent.)

and that each of the claims of the patent covers this function of twist arrest in this language:

“A false twisting device disposed to impart a false twist to the filaments emerging from said tube, whereby the twist feeds along said filaments into the tube and is set therein.” (Column 2, Lines 59-62 of the patent.)

But, plaintiffs’ argument continues, it is an undeniable fact that the ’375 heater does not arrest the twist therein and that this court, speaking through Judge Hemphill in an order dated November 14, 1973, has so held:

“It is established by the undisputed evidence and testimony that in the actual use of the ARCT FT machines . there is no twist arresting device nor operation which stops the twist in the heater nor is there any heating of a length of the yarn in the heater in an untwisted condition. On the contrary, there is undisputed evidence and testimony that in the actual use of these ARCT FT machines . . . the twist which feeds backwardly from the false twist device is not stopped or arrested in the heater, but travels entirely through the heater and on back to the yarn feed rolls, and that the yarn is not heated in an untwisted condition in the heater.” Duplan Corporation v. Deering Milliken, Inc., 370 F.Supp. 769, 779 (D.S.C.1973).

Plaintiffs further contend that their argument is bolstered by reference to the file wrapper which discloses that following the first rejection of the patent the solicitor for Chavanoz argued that the feature of twist arrest within the heater distinguished the ’375 patent over the prior art cited against it by the examiner in his initial denial.

Plaintiffs therefore conclude that the only compatible construction of the claims, specifications and file wrapper is that the patent claims an apparatus in which the twist is arrested in the heater and that since in fact no ARCT machine involved arrests twist in the heater, the ’375 patent, even if valid, is not infringed.

*726 The defendants in a three-point rejoinder argue: First, that the necessity for equipping a false twist machine with some means of stopping the twist before allowing it to reach the supply package was well known in the prior art cited in the ’375 patent; that the failure to include such an element in the drawing of the patent is attributable to the fact that this is merely a schematic illustration focusing mainly on the heating element; and that any person skilled in the art would automatically know to add the appropriate means for stopping the twist; second, that the examiner did not rely on the statements of the solicitor is demonstrated by his rejection of the ’375 application a second time and the failure of the solicitor to press the twist arrest feature in his second set of remarks that immediately preceded the allowance of the patent; and finally, defendants argue that in any event arrest of the twist does not necessarily mean to stop it entirely, and that therefore any marginal arrest of the twist that might occur in the heater, as some experts believe it does, would satisfy the specification of the patent.

The court has concluded that the plaintiffs have the better end of the argument and that they have not infringed the ’375 patent.

Essential to the operation of every false twist machine is some means for arresting the twist between the heater and the supply package. Without such means the twist will migrate backward into the supply package causing yarn entanglement, loss of control over the false twist process and eventual breakage. Witnesses for both sides testified to these facts. Also undisputed is the fact that the only means disclosed by the ’375 patent for arresting the twist was by “contact of the [yarn] with the surface of such [curved] tube.” The claims do not spell this out with the same specificity, using instead the language, “the twist feeds along said filaments into the tube and is set therein.”

In distinguishing the Ancet ’Oil patent the solicitor for Chavanoz wrote that the contact achieved by the curbed tube heater “ . . . between the filaments and the walls serves as a means for arresting the twist so that the twist can be confined to the zone of the filament between the false twister and the point of contact with the heated tube.” (PX 49 at pp. 10-11.) (Emphasis added.)

and in distinguishing the Stoddard ’105 patent he said:

“[I]n the Stoddard construction there is no contact between the yarn and the walls of the heated tube. Hence there is no means for arresting the twist and confining it to the limited length of yarn between the twister and the end of the tube as in the applicants’ case.

******

“[B]y arresting the twist by contacting the tube the filaments are heated while in untwisted condition. This facilitates heat transfer to the multiple filaments.” (PX 49 at pp. 11-12.) (Emphasis added.)

It is settled law, of course, that the claims must be construed in the light of the specifications and file wrapper. Graham v. John Deere Company, 383 U.S. 1, 33, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966); Schriber-Schroth Company v. Cleveland Trust Company, 311 U.S. 211, 61 S.Ct. 235, 85 L.Ed. 132 (1940). When so construed the specifications and claims of the ’375 patent together with the file history and the arguments advanced during the course of its prosecution in the Patent Office establish beyond peradventure that an essential feature of the invention and one on which its operability was dependent was its capacity for arresting the twist within the heater. As we have seen, on the ARCT machines operated by the plaintiffs there is “no twist arresting device which stops the twist in the heater.” 370 F.Supp. at p. 779.

What, then, is the effect of this failure of the invention to perform as advertised? The answer is found in a plethora of cases which hold that even where an “unambiguous” claim may literally read on an accused machine, there may nevertheless *727 be no infringement if there is not real identity of means, operation and result. 61

“[TJhere will be no infringement, though the claims of a patent literally happen to read on some physical embodiment, where such physical embodiment does not employ substantially the same means, in substantially the same way, to achieve substantially the same results.as that claimed. This rule, sometimes referred to as the inverse doctrine of equivalents, says, in effect, that an incidental infringement is no infringement at all. Westinghouse v. Boyden Power Brake Company, 170 U.S. 537 [, 18 S.Ct. 707, 42 L.Ed. 1136] (1898).” Rosenberg, “Patent Law Fundamentals”, p. 302 (Clark Boardman Company, Ltd., 1975).

Here the claims of the ’375 patent literally read on the ARCT machines, but these machines employ a means not present in the patent (feed rolls disposed between the heater and the supply package) to perform the essential function of twist arrest which the patent erroneously claims is performed in the heater itself.

It may be, as defendants argue, that any person skilled in the art would know that he would have to devise some way to arrest the twist in the yarn before it migrated back to the supply package, but the argument is incongruous, for the patent on its face discloses that the twist will be arrested in the heater and that nothing more is required.

In connection with their argument that the examiner did not rely on the solicitor’s twist arrest representations defendants complain that the plaintiffs did not produce the examiner as a witness at the trial. But defendants overlook the fact that they had the burden of proof on the issue of infringement, and if they wanted the testimony of the examiner, they should have produced him. At this point the court cannot speculate as to what the examiner had in mind in allowing this patent. Suffice it to say that he had before him the solemn representations of defendants’ solicitor, the claims and the specifications to the effect that the indispensable function of twist arrest was accomplished in this invention by the contact of the running yarn against the inner surface of the curved tube heater — representations which are now known not to have been true — and that the patent was allowed while these representations remained uncorrected.

Defendants’ final argument to the effect that “twist arrest” as employed in the ’375 patent means “partial arrest” or something less than complete twist stoppage must also be rejected. In his order previously referred to Judge Hemphill held that:

“[T]he patent owner is estopped from giving the term of art, ‘arrest the twist’, in the patent claims any different meaning than that which appears incontrovertibly clear in the file wrapper. Therefore, this court concludes that there is no genuine issue of fact that the term ‘arrest the twist’ means ‘arrest all the twist’ and not ‘arrest most of the twist’.” 370 F.Supp. 769 at p. 787.

In summary, even if the ’375 patent were valid, it is not infringed by the plaintiffs’ use of the ARCT machines in which the invention is embodied. This is so because the patent was issued on a false twist apparatus having a curved tube heater which arrests the twist. The heaters on the machines used by the plaintiffs do not do this, and the patent is therefore not infringed.

In addition to the facts found by the court in its treatment of the. ’375 patent the *728 court adopts as its own the following proposed findings of fact submitted by the parties on the patent issues:

1. On the question of anticipation: Plaintiffs’ proposed findings 6.3.1 — 6.3.4; 6.3.5 with the exception of the third and fifth sentences; 6.3.7; and 6.3.9.

2. On the question of the application of the Lincoln Engineering doctrine: Plaintiffs’ proposed findings 6.4.1; 6.4.2; 6.4.4; and 6.4.5.

3. On the question of obviousness: Plaintiffs’ proposed findings 6.1.1; 6.5.1-^6.-5.4; 6.6.3; 6.6.4; 6.6.6; 6.6.7 except for the last sentence; the first two sentences of 6.7.1; and 6.7.2-6.7.4; and defendants’ proposed findings 2; the first three sentences of No. 7; 10; 55-57; 77; 80; and 87.

4. On the question of infringement, plaintiffs’ proposed findings 6.2.1-6.2.6.

Patent No. 2,944,319

The title of this ’319 patent is a “Heating Device for Filaments”. It was issued to Chavanoz as assignee of the inventor, Henri Crouzet, on July 12, 1960. The invention consists of a removable metal tube of approximately three inches in length which is telescoped into a flanged “sleeve” at the top of the heater tube on a false twist texturing machine. The purpose of the invention is stated in the specifications as follows:

“When synthetic filaments are subjected to high-temperature heat treatment, vapors are generally evolved which usually condense near the top of the heating element at a point where the temperature begins to drop off. This condensation gives rise to a solid deposit which must be removed from time to time.

“With a view to overcoming this drawback, the invention provides a heating element designed for the treatment of textile filaments, and specifically of superpolyamide filaments, which consists of a two-part tubular element. The part located on the side where the filament enters is permanently attached to the machine and constitutes the heating element as such; whereas the part on the side where the filament leaves, located in the temperature region where the vapors condense, is removably mounted.”

There follows a reference to the two drawings reproduced below:

The specifications continue:

“In the drawing:

“Fig. 1 is a longitudinal section of a heating element of the conventional type; and

“Fig. 2 is a similar view of a heating element embodying the invention.

“Referring first to Fig. 1, a heating element of the conventional type consists of a tube 2 fitted with terminals 11 permitting it to be connected to an electrical circuit for heating purposes.

“In the unheated part of the tube 2, that is, in the region where the temperature begins to fall off, solid deposits 4 form, caused by condensation of the vapors which are evolved in the course of the heat treatment of the filament passing through the tube.

“Fig. 2 shows a heating element embodying the invention. Said element consists of a lower tube 5 carrying at its upper end a conical sleeve 6 into which there is removably set a short tube 7 that forms an extension of tube 5 and is held in place in the upper part of said conical *729 sleeve by an insulating ring 8. Terminals 12 located on the lower part of tube 5 and also near the upper conical sleeve 6 permit the heating element to be connected to a source of heating current.

“As may be seen from Fig. 2, the deposits due to condensation of the vapors that are evolved during the heat-treatment of the filament which passes through the heating element will form in the upper tube 7 on the level of line A-A (sic, ostensibly “A — B”), in the region where the temperature drops.”

And the virtue ascribed to the invention is stated thus:

“In view of the ease which tube 7 can be removed, it is evident that the downtime of the machine for the removal of deposits 10 caused by condensation is reduced to a minimum. Actually, all that need be done is to remove tube 7 of each element and replace it with an identical tube free of deposits, which only takes a moment. The cleaning of the encrusted tube 7 is then done outside of the machine, in a caustic soda bath, for example.”

From the drawing shown above it will be observed that condensation of vapors generated by the texturing process takes place at the same point and in the same amount in each of the two figures. It is evident, therefore, that the real problem sought to be solved by this invention initially was the difficulty in cleaning the top portion of the heater tubes where residue from the heated yarn vapors inevitably collected. As stated in the specifications, if allowed to collect in the removable tube insert, obviously this residue could be more easily and efficiently removed.

Originally the application contained three claims all of which were rejected by the examiner. Only in claim 3 was there mention of “insulating means”. That claim stated that the sleeve holding the removable tube should have “insulating means adapted to support said second tube.” But no particular virtue was attributed to “insulating means” except as it functioned to provide support for the removable tube insert.

In rejecting all three claims the examiner, citing three patent references, stated that removable tubes were well known in the prior art and that “it appears obvious that, should condensation products form in these extended sections, they could be removed and cleaned without exercising inventive reasoning.”

The solicitor apparently acquiesced in this rejection, for he cancelled the original three claims and applied for a patent on a new claim which is the one that ultimately issued. That claim reads as follows:

“Apparatus for the treatment of textile filaments comprising a fixed tube having a substantially vertical axis, means heating said tube to a temperature suited for the treatment of said filaments, a sleeve disposed at the upper end of said tube and a second tube removably mounted in said sleeve in axial alignment with said first tube, said sleeve having heat-insulating means to retard the transfer of heat from said first tube to said second tube whereby condensation of vapors rising from said first tube may take place on the walls of said second tube, and means feeding a textile filament for treatment upwardly through said tubes.”

The difference in the cancelled claims and the claim that issued as the ’319 patent is the requirement that the sleeve holding the removable tube have “heat-insulating means to retard the transfer of heat from said first tube to said second tube”, and in prosecuting this new claim the solicitor undertook to distinguish the prior art cited by the examiner by pointing out that the removable tubes in the prior art patents did not disclose heat insulation. He further represented that without such insulation “between the tubes” cooling and condensation of vapors in the removable tube would not take place.

Shortly thereafter the patent was granted without further ado, and the inference is inescapable that the shift of emphasis in the prosecution of the patent from the claimed improved accessibility of the removable tube for cleaning purposes to the advan *730 tages ascribed to the heat insulating means in retarding the transfer of heat between the two tubes tipped the scales in favor of patentability. All of this was accomplished without material alteration or amendment of the original specifications.

Plaintiffs have mounted a three-pronged attack on the ’319 patent. They argue first, that it is void for obviousness; second, that they have not infringed the patent; and third, that in any event it is unenforceable because certain statements made by the solicitor to the Patent Office concerning the effect of the heat transfer retardation attributable to the insulating means were admittedly false.

1. Obviousness. In support of their argument on obviousness the plaintiffs do not rely on any prior art other than that disclosed in the file wrapper of the patent (PX 56). Thus the only Section 103 question left for decision is whether in view of simple, universally known scientific principles the subject matter of the ’319 patent “would have been obvious at the time the invention was made to a person having ordinary skill in the art.” The court has concluded that it would and that the patent is void for obviousness. 62

At the time this patent issued it was well known that condensation occurred in the top of heater tubes on false twist machines because the top, being exposed to the open air, was necessarily cooler. And, of course, it is universally known that insulating an object from a heat source serves to keep it cooler. Given these two simple, indisputable facts, the court has no difficulty in holding that any man of ordinary skill in the art would have known by insulating the removable insert tube its efficiency as a condenser could easily have been enhanced. 63

2. Infringement. On the issue of infringement plaintiffs argue initially that on the machines they use there is no insulation at all between the bottom of the insert tube and the heater tube proper at the place they come together. 64 Thus they say that the portion of the patent’s claim which states “a sleeve disposed at the upper end of said tube and a second tube removably mounted in said sleeve in axial alignment with said first tube, said sleeve having heat-insulating means to retard the transfer of heat from said first tube to said second tube” does not read on their machines.

Defendants counter with the argument that the sleeve referred to in the claim is part and parcel of the first tube referred to; that on all ARCT machines there is a one-half inch T-shaped rubber or bakelite grommet which fits in the top of the sleeve and through which the removable tube is inserted; and that this grommet serves as the “heat insulation means” required by the claim. 65 Additionally, defendants contend that the air space between the inside of the sleeve and the outside of the insert tube serves as a “heat insulating means” satisfying the claim’s requirements.

To all this the plaintiffs rejoin that the grommet serves no purpose other than to *731 stabilize the insert tube within the sleeve and to keep it from flapping against the inner walls of the sleeve; that it has no heat insulating qualities at all; and that nowhere in the patent is there any mention of the air space as a heat-insulating means.

The court is of opinion that the defendants are correct in their assertion that the sleeve is a part of the heater tube and that the admitted fact that there is no insulation at the point where the bottom of the insert tube comes in contact with the heater tube cannot avail the plaintiffs. At the same time the court rejects the defendants’ argument that the air space between the sleeve and the insert tube is a heat-insulating means, this argument being, in the court’s opinion, a mere afterthought nowhere hinted at in the patent itself.

This leaves for consideration the question of whether the grommet constitutes a “heat insulating means”. While plaintiffs had an expert to conduct an experiment which showed there was no appreciable difference between the temperature of the heater tube below the insert tube and the insert tube itself, neither party apparently conducted any tests to show temperature comparisons between the sleeve and the insert tube, and the record is without technical evidence as to whether the insert tube is actually cooler than the sleeve with the grommet in place than it is without it. 66

Common sense would seem to dictate that if there is any such difference, it is de minimis. It is just not reasonable that this hard rubber ring extending not more than a half inch down into the sleeve could impede by any appreciable amount heat transfer between the sleeve and the insert tube sufficient to cause increased condensation of the vapors at that point. In fact, the patent itself describes the function of this “insulating ring” as that of holding the insert tube in place in the conical sleeve.

The defendants have not sustained their burden of proving infringement of the ’319 patent by the parties so charged.

3. Enforceability of the ’319 Patent. Even if the patent were held valid and infringed, plaintiffs contend that because of an admitted misrepresentation made to the Patent Office by the solicitor for the inventor in his response to the rejection of all the claims in the patent as originally filed, the patent is not enforceable under controlling equitable principles. The statements in question appear at page 10 of the file wrapper (PX 56) and read as follows:

“The Ferre and Dickie patents show an apparatus for drying a coating on a wire. These patents do not disclose apparatus having a structure defined in the present claim wherein a removable upper tube is mounted in a sleeve or socket at the end of a heating tube and is heat-insulated. Without this heat-insulation between the tubes, the cooling and condensation of the vapors would not take place prior to their discharge into the air. The Ferre and Dickie apparatus shows no means for controlling the relative heating and cooling as above stated. (Underlining added.)

“Likewise the Kunzle patent fails to show a combination of heated and unheated tubes with the unheated tube heat-insulated and removable as defined in the present claims. While Kunzle discloses a pair of aligned tubes, there is no control of the heating and cooling to insure that the vapors will deposit on the walls of the second tube. This requires the combination of a heated first tube and a heat-insulated second tube.”

In his deposition testimony the inventor, Crouzet, after first acknowledging that the portion of the statement underlined above was false (Tr. 5, p. 961), later in his response to a question by his own counsel *732 modified his testimony to say that this statement was “inaccurate” (Tr. 5, p. 965).

Defendants contend that the solicitor’s statement should not be given its clear and obvious meaning for that the drawings and specifications already on file disclose that vapors regularly condensed in the upper portions of the heaters without the aid of the invention of this patent. They point out that the first paragraph of the specifications states that “vapors are generally evolved which usually condense near the top of the heating element at a point where the temperature begins to drop off”. Furthermore, the drawing submitted with the original application show the accumulation of condensed vapors in the upper portion of a one-piece, tubular heater. Therefore, they assert that the statements of the solicitor must be interpreted as meaning only that this invention would cause additional condensation over that which would normally be expected to occur.

Although this appears to be another example of a patent solicitor’s willingness in pre-Lear v. Atkins 67 days to say whatever was necessary to obtain allowance of a patent, it is difficult to conclude from the disclosures in the overall application that the examiner was in fact misled. Presumably he was familiar with the proposed specifications and drawings and would have interpreted the remarks of the solicitor in their light. Penn Yan Boats, Inc. v. Sea Lark Boats, Inc., 359 F.Supp. 948, 955 (S.D. Fla.1972). While the court believes the representations were intentionally made, it cannot be concluded on this evidence that the patent would not have issued except for these misstatements. Therefore, the patent cannot be invalidated for fraud in its procurement.

But the law is that

“Even though misrepresentations made to the Patent Office are not legally material to the issuance of a patent, nevertheless, this Court, being a court of equity, can and should refuse to enforce the patent if the Court finds the patentee made intentional misrepresentations to the patent examiner, i. e, if the patentee came into the court with unclean hands. The proceeding before the patent examiner is ex parte and an examiner has no way, in many cases, to ascertain the truthfulness of the representations made to him. Necessarily he must rely on the good faith of the applicant. Absolute honesty and good faith disclosure is necessary.” Corning Glass Works v. Anchor Hocking Glass Corporation, 253 F.Supp. 461, 470 (D.Del.1966), reversed on other grounds, 374 F.2d 473 (3rd Cir. 1967), cert. denied, 389 U.S. 826, 88 S.Ct. 65, 19 L.Ed.2d 80 (1967).

Application of these principles, which have been widely accepted by the courts, see Precision Instrument Manufacturing Company v. Automotive Maintenance Machinery Company, 324 U.S. 806, 65 S.Ct. 993, 89 L.Ed. 1381 (1945); SCM Corporation v. Radio Corporation of America, 318 F.Supp. 433, 449 (S.D.N.Y.1970); Jack Winter, Inc. v. Koratron Company, Inc., 375 F.Supp. 1, 49 (N.D.Cal.1974); and Penn Yan Boats, supra, at p. 968, has led the court to the conclusion that regardless of its validity or infringement, the ’319 patent is not enforceable. Whether or not the examiner was actually misled by the statement, the inventor admitted that it was not true, and at best the court finds that it was made with careless disregard for its accuracy. Furthermore, the statement was not made with regard to some minor advantage of the invention but instead was the only argument advanced by the solicitor to distinguish the ’319 invention from the prior art. Any doubt as to whether the examiner was misled should be resolved against the inventor. SCM Corporation, supra.

In summary, the court holds that the ’319 patent is invalid for obviousness; that even if valid the plaintiffs, Burlington and Schwartzenbach-Huber, have not infringed it; and that in any event the patent is not enforceable because of the admitted material misrepresentations made by the inventor’s attorney to the Patent Office.

*733 In addition to the facts found by the court in its treatment of the ’319 patent the court adopts as its own the following proposed findings of fact submitted by the parties on the patent issues:

1. On the issue of obviousness: plaintiffs’ proposed finding of fact 7.4.3; defendants’ proposed findings of fact 1-20, 23, 26, 27 and 29.

2. On the issue of infringement: plaintiffs’ proposed findings of fact 7.3.1 and 7.3.3 (last two sentences); defendants’ proposed findings of fact 33, 34, 36-38, 43, 44 and 46-54.

3. On the issue of enforceability: plaintiffs’ proposed findings of fact 7.1.1 and 7.2.3-7.2.5.

Patent No. 3,012,397

United States Patent No. 3,012,397 (Polyester Process) was issued to Chavanoz on December 12, 1961 upon application filed on November 30, 1959 by Henri Servage, then a Chavanoz employee. The patent asserts fourteen claims for methods for producing high bulk polyacrylonitrile, acrylonitrile co-polymer, and polyester yarns. Only claims 9 through 11 are at issue in this case.

Claim 9 involves the following steps: (a) false twisting and permitting filament heat shrinkage of the yarn; (b) overfeeding and “mechanically spatially shrinking” the yarn 5 to 12% subsequent to the first heat shrinkage; and (c) then subjecting the yarn to a subsequent heat treatment while under “substantially no appreciable tension.” Claim 10 relates to claim 9 when the filament heat shrinkage during the false twist step is in the range of 0.5 to 2%. Claim 11 relates to claim 9 when the yarn is subjected to a subsequent heat treatment in a continuous or double heater process.

As discussed in the specifications, the ’397 patent relates to a process for texturing continuous filament yarns from polyester and acrylic fibers to produce what is known as a “high bulk” textured product by means of an additional heating of the yarn after the false twist step. Fundamentally, the process consists of a specific range of combined conditions to be effected during the false-twisting and subsequent heating steps. It instructs one to work in this range to get the sought-after results, namely bulk, defined as curl and increased volume.

The first condition to be observed is to permit filament heat shrinkage of the yarn, for example on the order of 0.5 to 2% during the- false-twisting step. After false-twisting the yarn is next put into a condition where it experiences a mechanical shrinkage of 5 to 12% and is given a second heat treatment, which treatment is in accordance with the teachings of the French Third Patent of Addition No. 67,751 (PX 637 — 1799). The second heat treatment is carried out either on the take-up package in a later step (a discontinuous process) or on a double heater false twist machine as the yarn proceeds to the take-up package through a second heater (a continuous process). Comprehending each of these alternatives, the patent specifies overfeeding 68 the yarn under controlled conditions, specifically 5 to 12% mechanical shrinkage, and giving it a second heat treatment while the yarn is under “substantially no appreciable tension.” Plaintiff Throwsters have asserted that this patent is not infringed, is invalid for lack of novelty or for obviousness (35 U.S.C. §§ 102 and 103), and is invalid for indefiniteness (35 U.S.C. § 112). Defendants, of course, assert that the patent is both valid and infringed. Each position will be discussed below.

Infringement of the ’397 Patent.

For purposes of deciding the infringement issue, the court will assume the validity of the ’397 patent. The burden of proof of infringement is on the defendants (patentees and licensors).

*734 Plaintiff Throwsters’ essential argument on non-infringement focuses on the fact that the ’397 patent requires a second heat treatment while the yarn is in a state of “substantially no appreciable tension.” The Throwsters submit that they have processed polyester at overfeeds of from 10 to 16% in the second heater which has produced tension on the yarn of from 2 to 3 grams. They assert that this tension is low, but is nonetheless appreciable and tends to elongate or straighten the yarn somewhat while it is being heated. Thus, the resultant yarn would display characteristics of reduced curl and stretch rather than the increased curl or bulk as promised by the ’397 patent.

The phrase “substantially no appreciable tension” is not defined specifically in the patent. If a discontinuous process is utilized, the patent indicates that the requisite state of tension can be obtained by heating the yarn while in skein form by placing the skeins in an oven at moderate temperature. If a continuous process is utilized, the patent simply states that the second heat treatment should be under no appreciable tension “as by overfeed of the yarn through this heating zone.”

The tension of the yarn as it is subjected to a second heat treatment is in each case determined by the amount of overfeed. The higher the overfeed, the lower the tension on the yarn. Plaintiffs have asserted that the “no appreciable tension” requirement requires overfeeds substantially above the 10 to 16% overfeeds practiced in their specifications and that the tension should be closer to zero than the 2 to 3 grams tension obtained in their processes. Defendants simply say that the disputed phrase is a common sense term and that one simply overfeeds in the second heat zone to obtain a low rate of tension such as the 2 to 3-gram figure mentioned by plaintiffs.

The court concludes that a fair reading of the patent would not require one of ordinary skill in the art to overfeed the yarn at rates substantially above those run by the plaintiffs. At various points the patent speaks in terms of overfeeds and shrinkages of from 5 to 12% and there is no way that overfeeds of this magnitude could yield the “approaching zero” gram tensions that plaintiffs indicate the patent requires. A fair reading of the patent requires one to overfeed so as to obtain a low tension, substantially less than the tension in the first heating and false-twisting zone. Accordingly, plaintiffs’ argument of non-infringement is rejected and evidence that the Throwsters ran processes yielding from 2 to 3 grams of tension establishes infringement.

An alternative argument made by plaintiffs for non-infringement is more appropriately considered as an attack on the patent’s validity. Essentially, plaintiffs argue that the processes they used were those disclosed in the prior art and not those taught in the ’397 patent. They specifically deny their infringement in utilizing defendants’ “novel” step of mechanical spatial shrinkage. For present purposes, the validity of the ’397 patent is assumed. Accordingly, if valid, the defendants have established that processes run by the plaintiffs in treating polyester yarns have infringed the ’397 patent.

Validity: Anticipation or Obviousness.

In the course of this trial approximately forty patents were cited to the court as prior art to be considered in evaluating the validity of the ’397 patent. A strong case has been presented that several of these patents, particularly Scragg United States Patent No. 3,472,011 and Klein and Evans United States Patent No. 2,977,746, anticipated the ’397 patent. Anticipation, however, is a narrow, technical defense, requiring a showing of virtual identity between the patent in question and a specific prior art device. E. g. Diversified Products Corporation v. Sport Stores, 294 F.Supp. 375 (D.C.Md.1968). By comparison, obviousness is intended to be a demanding standard where the court looks at the whole of the prior art in considering it together for all that was disclosed therein. Graham v. John Deere Company, 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). The court has determined that it is not necessary to decide *735 the issue of anticipation under Section 102 since the ’397 patent is invalid for obviousness in light of the teachings of the prior art.

As we have seen, in evaluating obviousness under Section 103 the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be entertained, and the level of ordinary skill in the pertinent art is to be resolved. In resolving the level of skill inquiry, the court finds that in the fall of 1959, it was within the skill of persons of ordinary skill in the art of processing thermoplastic yarns to vary the amount of overfeed through the second heater, depending on the nature and denier of the yarns being processed, in order to obtain the desired characteristics of the end product. There was a great interest in discovering an effective method for processing set polyester. This natural progression in the art resulted in a number of patent applications being filed by Heberlein, Chavanoz, DMRC, Leesona and Scragg in the mid to late 1950’s all relating to post-treatment of crimped yarns by overfeeding and heating under controlled conditions.

The court will not analyze all the prior art cited against the ’397 patent, but will summarize the more relevant prior art and adopt proposed findings as to the remainder.

The three most relevant patents contained in the prior art were Scragg United States Patent No. 3,472,011; Klein and Evans (DMRC) United States Patent No. 2,977,746; and Seem and Stoddard Australian Patent No. 219,270. The Scragg ’Oil patent was filed in July of 1958 and corresponds to French Patent No. 1,199,764 which was available to the public on July 23, 1959. It covers a double heater false-twisting system for processing thermoplastic yarns. The patent discloses overfeeding yarn through the false twist zone of the first heater to permit shrinkage of the yarn, a shrinkage of 2% being indicated as suitable. It also teaches overfeeding in the second heat zone, said heat zone being at a lower temperature than the first, and specifically suggests overfeeding on the order of 20% or less.

The Klein and Evans Patent No. 2,977,-746 was filed in June of 1958 and issued on April 4, 1961. It is a DMRC patent dealing with a method for thermally treating thermoplastic yarns and post-treating elasticized yarns, with polyester and nylon being specifically mentioned. Although the patent is concerned primarily with an edge-crimping process, interestingly it discloses that the elastic nature can be imparted to thermoplastic yarns by several well-known processes including uptwisting or downtwisting or using a false-twisting apparatus. 69

Australian Patent No. 219,270 was published on May 29,1958, over a year prior to the filing of the ’397 patent, and discloses apparatus and methods for re-processing torque stretched yarns by the false twist process. The ’270 patent teaches that tension on the yarn as it passes through the first heater of a false twist machine can be controlled by underfeeding or overfeeding the yarn. This patent has been described as an encyclopedia on false-twisting yarns by a second heater treatment. It teaches that characteristics of yarn may be modified by regulating the tension on the yarn after it has been false-twisted or by subjecting it to a second heat treatment or by a combination of both. According to the patent, tension is a function of overfeed and a full range of tensions can be obtained by overfeeding in the second heat treatment, which may be applied continuously or discontinuously. Finally, the ’270 teaches that a wide range of end products (varying in degree of stretch or curl or bulk) can be obtained by varying the tension of the yarn as it is subjected to the second heat treatment.

*736 Various other items of prior art have been presented to the court. These include United States Patent No. 2,761,272 (Chavanoz), disclosing that transmission ratios between the feeding and take-up devices may be selected to permit any desired shrinkage; United States Patent No. 2,987,869 (DMRC) disclosing that elasticity or activity of yarns can be increased by low tension heat treatment; United States Patent No. 3,181,528 (Scragg) disclosing overfeeding the yarn (permitting filament shrinkage) in the false twist zone and overfeeding through a second heater at a rate of 30 to 40%; United States Patent No. 3,017,685 (Heberlein) disclosing a continuous second heater process with the use of low tension; .05 to .2 grams in the second heater; and United States Patent No. 2,919,534 (DMRC) disclosing apparatus and processes for subjecting crimped yarns (both twist and edge-crimped) to a subsequent heat treatment. Finally, the prior art also included commercial marketing by Leesona in 1957 through 1959 of a 511 attachment for post-treating yarns by the SAABA process. Although the SAABA process called for stretching rather than overfeeding the yarn in the first heater, the process did call for overfeeds through the second heater.

None of the foregoing prior art was cited to the patent examiner during the prosecution of the ’397 patent. The court finds this art to be more pertinent than the art disclosed or cited by the patent examiner in the prosecution of the ’397 patent. According to the law of this Circuit, the presumption of the validity of a patent is weakened or destroyed where there has been a failure to consider and to cite the most pertinent prior art. Heyl & Patterson v. McDowell, 317 F.2d 719 (4th Cir. 1963).

Defendants attempt to distinguish the ’397 process patent from the foregoing prior art. It is argued that the ’397 is not a patent for overfeeding or for giving a second heat treatment. Instead, it is asserted to be a narrow process patent teaching one to permit filament heat shrinkage in the first heater zone, and then to put the yarn in a condition of 5 to 12% mechanical spatial shrinkage and to set it there. Defendants’ position regarding the concept of “mechanical spatial” shrinkage is central to their argument on patent validity. This concept will be more fully discussed later in the section on indefiniteness. For present purposes, it is sufficient to note that yarn shrinkage is inherent when one overfeeds a previously false-twisted yarn through a heated zone. Total shrinkage is directly related to the amount of overfeed. In fact, in the discontinuous process, defendants admit that “mechanical” shrinkage in the yarn is equal to the net overfeed onto the take-up package.

Prior art had disclosed both overfeeding in the false-twisting zone and overfeeding prior to or during a second heat treatment of polyester or other thermoplastic yarns. Defendants attempt to distinguish this prior art, and particularly the Australian ’270 “encyclopedia” patent, as being so broad as to be worthless; whereas, the ’397 patent allegedly teaches quickly how to obtain high bulk yarn. Had the '270 patent comprised the whole of the relevant prior art, the court would be more sympathetic to this argument. The art, however, was not stagnant and continued to progress with further refinements of the optimum range of working conditions. The Scragg ’Oil patent and the Klein ’746 patent disclosed overfeeds of 20% or less in the second heater. Defendants’ expert, Mr. Platt, admitted on cross-examination that overfeeding on the order of 20% or less would produce mechanical or spatial shrinkage in the range of 5 to 12%. Defendants assert that the ’397 patent further minimized, although it did not eliminate, the trial and error necessary to obtain high bulk polyester yarns by narrowing the range of optimum working conditions from less than 20% overfeed to overfeeding and mechanically spatially shrinking in the range of 5 to 12%. Even assuming that the ’397 teachings were a further refinement on the prior art’s disclosures of less than 20% overfeed, the court finds that this was not a patentable refinement. It is not inventive to point out a particular range of conditions or optimum working ranges if what is involved is *737 nothing more than the skill of the mechanic and the exercise of “patient experimentation.” American Potato Dryers v. Peters, 184 F.2d 165 (4th Cir. 1950); Triumph Hosiery Mills v. Alamance Industries, 299 F.2d 793 (4th Cir. 1962). A patentee simply cannot designate a range in a known progression and maintain a monopoly on processes within that range on the ground that it produces optimum results.

The court’s inquiry must focus on whether a skilled artisan, with knowledge of the prior art, would require more than ordinary skill to discover the ’397 process. The court finds that it would not. Accordingly, the court holds that the subject matter of the claims of the ’397 patent would have been obvious to a person of ordinary skill in the art prior to November of 1959 in view of the teachings of the Australian Seem and Stoddard ’270 patent either alone or in combination with any one of the following:

(a) commercial use of the 511 attachment and the SAABA process disclosed in the ’270 patent;

(b) the Chavanoz Third Patent of Addition;

(c) the Scragg ’Oil patent;

(d) the disclosures of Messrs. Klein, Evans and Bolinger in any one of the several DMRC-Chavanoz patents and applications relating to post-treating thermoplastic yarns; and

(e) either Heberlein British Patent No. 710,802 or United States Patent No. 3,017,-685.

35 U.S.C. § 112: Indefiniteness.

Defendants have asserted at trial that the ’397 is not a patent concerned merely with overfeeding or with second heat treatments for polyester yarns. Central to defendants’ position on patent validity is the novel concept of mechanical spatial shrinkage. Claim 9 requires the step of overfeeding and mechanically spatially shrinking the yarn approximately 5 to 12%. In a discontinuous process, defendants admit that mechanical spatial shrinkage is equal to overfeed. However, in a continuous double heater process, the equation is more complicated since the yarn is shrinking both as a result of being relaxed and also as a result of the filament heat shrinkage. In this latter situation, defendants submit that mechanical spatial shrinkage is equal to overfeed minus heat shrinkage.

This definition of mechanical spatial shrinkage (“MS”) was developed at trial. The patent itself provides no definition of mechanical spatial shrinkage. The term overfeeding is a term commonly used in the art while the terminology “mechanical spatial shrinkage” is not. Defendants’ expert, Mr. Platt, admitted that no one in the art thought in terms of mechanical spatial shrinkage in 1959. The terminology “mechanical spatial shrinkage" does not appear in any of the thousands of documents, specifications or processing data produced in this litigation, other than the ’397 patent papers. There is no evidence that the art was swept into a tide of commercial production of polyester once the range of 5 to 12% mechanical spatial shrinkage was disclosed by the ’397 patent. Instead, as evidenced by the specifications and processing data produced at trial, Throwsters continued to think in terms of overfeeding and of modifying overfeeds, temperatures, and yarn deniers to get the end product yam characteristics that were sought.

The ’397 patent provides no definition of what mechanical spatial shrinkage is, how to determine it, how to measure it, or how to set up a machine to obtain it. At trial, no witness was able to determine whether a given processing specification would infringe the ’397 patent without resorting to a series of complex tests developed by defendants’ expert, Mr. Platt, which purported to delineate shrinkage resulting from relaxing the yarn by overfeeding from shrinkage resulting from heat.

Plaintiff Throwsters generally processed polyester yarns on ARCT double heater machines with common rates of overfeed from 10 to 16%, and more recently from 18 to 20%. In the absence of further definition and explanation in the patent, there was no reliable way for a Throwster to insure that *738 he was operating outside the patent range of 5 to 12% mechanical spatial shrinkage. Mr. Platt testified that a Throwster using prior art disclosures of overfeeds of 20% or less could have processing specifications which would infringe the ’397 patent. In fact, in two of Mr. Platt’s tests, Nos. 211 and 230, it was discovered that the same rate of overfeed in the first and second heat zones resulted in different MS calculations, one within and the other outside the patent range. Further, according to test 237, a Throwster running Dupont polyester yarn would have MS calculations outside the patent range. Yet, if the Throwsters switched to an Enka yarn and ran the machine with the same deniers, temperatures, and overfeeds, there would result an MS calculation within the patent range. The court agrees with the plaintiffs that the patent law does not require alleged infringers to play this type of Russian roulette.

The law on indefiniteness is explained in two Supreme Court opinions, United Carbon Company v. Binney Company, 317 U.S. 228, 63 S.Ct. 165, 87 L.Ed. 232 (1942), and General Electric Company v. Wabash Appliance Corporation, 304 U.S. 364, 58 S.Ct. 899, 82 L.Ed. 1402 (1938). In General Electric at page 369, 58 S.Ct. at page 901 the court stated:

“Patents, whether basic or for improvements must comply accurately and precisely with the statutory requirements as to claims of invention or discovery. The limits of a patent must be known for the protection of the patentee, the encouragement of the inventive genius of others and the assurance that the subject of the patent will be dedicated ultimately to the public. The statute seeks to guard against unreasonable advantages to the patentee and disadvantages to others arising from uncertainty as to their rights. The inventor must ‘inform the public during the life of the patent of the limits of the monopoly asserted, so that it may be known which features may be safely used or manufactured without a license and which may not.’ The claims ‘measure the invention’ . . . In a limited field the variant must be clearly defined.”

In United Carbon, supra, 317 U.S. at page 232, 63 S.Ct. at page 168, the court stated:

“The courts in determining the questions of invention and infringement brought to them by respondent, no less than the parties-litigant, need and may insist upon the precision enjoined by the statute. To sustain claims so indefinite as not to give the notice required by the statute would be in direct contravention of the public interest which Congress therein recognized and sought to protect.”

To the extent that the ’397 patent is distinguishable from the prior art on the basis of the novel disclosures relating to mechanical spatial shrinkage, the patent is void for indefiniteness. The patent fails to meet the standards of definiteness described by the Supreme Court in United Carbon and General Electric and established by 35 U.S.C. § 112. The patentee simply failed to inform the public of the limits of his monopoly with the result that those producing the yarns could never know which processes could be safely used without a license and which could not. The patent is void for indefiniteness under 35 U.S.C. § 112.

In addition to the facts found by the court in its treatment of the ’397 patent the court adopts as its own the following proposed findings of fact submitted by the parties on the patent issues:

1. On the issue of infringement: plaintiffs’ proposed findings 8.3.1-8.3.3 inclusive; 8.3.4 with the exception of the sixth sentence; 8.3.5; 8.3.6 with the exception of the first, second and ninth sentences; 8.3.7; and 8.3.8.

Defendants’ proposed findings Nos. X-98-101 inclusive and X-103-106 inclusive.

2. On the issue of validity: plaintiffs’ proposed findings 8.4.1; 8.5.1-8.5.5 inclusive; 8.5.6, with the exception of the statement in parenthesis; 8.5.7; 8.5.9; 8.5.10; 8.5.12; 8.5.13, with the exception of the last sentence; 8.5.14-8.5.17 inclusive; 8.5.19; *739 8.5.20; 8.5.22-8.5.27, 8.5.29, with the exception of the first sentence; 8.5.30, 8.5.31; and 8.5.36.

Defendants’ proposed findings X-l, X-2, X-6, X-10, X-12, X-13, X — 28 with the exception of the last two sentences, X-49 and X-94.

Patent No. 3,123,973

United States Patent No. 3,123,973 was issued on March 10, 1964, to Chavanoz as assignee of the inventor, Henri Servage, upon application filed November 30, 1959. No claim for priority based on any earlier filed foreign patent was asserted. The earliest date of invention to which the ’973 patent can be entitled is November 30,1959.

The ’973 patent is a process for texturing continuous filament cellulosic based yarns, particularly cellulose acetate yarns. The patent application had a long and contorted history spanning four and a half years during which the application was rejected on six occasions. The claims were continually narrowed to meet objections raised by the patent examiner and eventually six claims were allowed. Claim 4 of the ’973 patent, which refers to a process wherein the yarn is allowed to shrink by up to 6% as it is wound into a package, was formally disclaimed by Chavanoz pursuant to 35 U.S.C. § 261 after the commencement of trial. The remaining five claims either expressly or by reason of dependency require the following steps: (a) false twisting a yarn having a cellulosic base; (b) winding the yarn onto a package under conditions permitting the yarn to undergo “considerable shrinkage;” and (c) subjecting the yarn to a second heat treatment at the temperature of 75 to 85° centigrade in the presence of steam while the yarn is “substantially free from tension.”

This process patent has been attacked as invalid on several grounds, including lack of novelty, obviousness and indefiniteness. It is also asserted that, even if valid, the patent is not infringed. Each position will be considered.

Validity: Anticipation and Obviousness.

Plaintiffs have asserted that two prior art patents issued to Donald Finlayson anticipate the ’973 patent. The first Finlayson patent is British Patent No. 464,981 which was issued in 1937. The second is United States Patent No. 2,174,573 which was issued on October 3, 1939. Without discussing these patents extensively at this time, the court notes that neither of these prior art patents taught a heat treatment of acetate yarn after false twisting at a temperature of 75 to 85° centigrade in the presence of steam. As indicated previously, the courts have construed anticipation as a narrow defense and, since the Finlayson patents do not specifically disclose the optimum range of temperatures disclosed in ’973, there is no strict anticipation. See National Lead v. Western Lead, 324 F.2d 539 (9th Cir. 1963).

Even though there is no strict anticipation, the Finlayson patents, when considered with the other prior art, may render the patent void for obviousness. The scope and content of the prior art must therefore be determined, the differences between the prior art and the patent claims must be entertained, and the level of ordinary skill in the art must be resolved.

A prime source of knowledge of the prior art is found in the file wrapper of the ’973 patent. Like so many of the Chavanoz patents, this one had a tortuous file wrapper history covering four and a half years during which the solicitor had three different interviews with the examiner and during which the claims were rejected no less than six different times. Each rejection was responded to and the claims were eventually amended and gradually narrowed down to the point where they were allowed.

During the prosecution of the patent, fourteen different prior art patents were cited by the examiner. The solicitor admitted that at least one prior art patent cited by the examiner disclosed rewinding with “considerable shrinkage” while others disclosed a second heat treatment; nonetheless, the solicitor submitted that it would “involve invention to provide a process in *740 volving both a winding technique under conditions, namely overfeeding, whereby considerable shrinkage would occur, followed by a heat treatment . . . ” (File Wrapper, PX 671, p. 47). Throughout the file wrapper the concept of considerable shrinkage was used to distinguish the invention from prior patents which taught only “low tension rewinding.” Finally, the prior art for processing cellulosic yarns was distinguished as failing to disclose a second thermal treatment at a moderate temperature and the range of 75 to 85° centigrade was specifically mentioned. Although a prior Heberlein patent cited by the examiner disclosed a range of between 60-200° centigrade the solicitor insisted that reading Heberlein in conjunction with Stucki United States Patent No. 2,869,318 would require one of ordinary skill in the art to conduct the second heat treatment at a high temperature, one in excess of 150° centigrade (PX 671, p. 60).

There were eventually three distinctions made between the ’973 patent invention and the prior art cited by the patent examiner. First, the prior art failed to disclose the combination of two prior processes: (1) a process with a winding technique allowing considerable shrinkage; and (2) a process involving a second heat treatment. Second, the prior art taught low tension rewinding rather than allowing considerable shrinkage to take place by overfeeding. Third, the prior art did not disclose a moderate second heat treatment in the presence of steam at a temperature of from 75 to 85° centigrade.

The two Finlayson patents, British Patent No. 464,981 (PX 14) and United States Patent No. 2,174,573 (PX 742) were not disclosed to or cited by the patent examiner. Taken together, these two patents disclose a method for improving cellulose acetate yarns which combines the steps of false-twisting, overfeeding to the take-up package to permit considerable shrinkage, and applying a second heat treatment with a hot aqueous medium (including wet steam) at a temperature at or close to the boiling temperature. The British patent specifically called for false twisting and overfeeding to the take-up package to permit shrinkage of the yarn. This British patent is incorporated in the later United States patent. The United States Finlayson patent specifically discloses post heat treatment of cellulose acetate yarns in the presence of steam. These two prior art patents were more pertinent than any of the art cited by the examiner in the prosecution of the ’973 patent. The presumption of the patent’s validity is weakened or destroyed because of the failure of the examiner to consider the most pertinent prior art. Heyl & Patterson v. McDowell, 317 F.2d 719 (4th Cir. 1963).

Defendants make several attempts to distinguish the Finlayson patents. They assert that the Finlayson patents were “paper patents” which were never practiced commercially and, accordingly, that they should be narrowly construed. Defendants also assert that the Finlayson patents are distinguishable in that they did not disclose the claimed range of from 75 to 85° centigrade in the second heat treatment.

In support of their position on the “paper patent” status of Finlayson, defendants point out that the Finlayson United States Patent No. 2,174,573 was in fact the patent of the Celanese Corporation. Defendants submit that attempts by Celanese to develop a process to texture cellulose acetate continued well into the 1950’s and are exemplified by United States Patent No. 2,869,318 which was filed in 1954. This latter Celanese patent did not utilize the step of post-treating cellulose yarn. Defendants assert that this patent is more indicative of the thinking of Celanese and of the thinking in the industry in the mid to late 1950’s than is the 1939 Celanese patent.

A patent constitutes prior art for 35 U.S.C. § 103 purposes even though it is a “paper patent” which was never put to commercial use. National Filters v. Research Products Corporation, 384 F.2d 516 (5th Cir. 1967); Sutter Products Company v. Pettibone Mulliken Corporation, 428 F.2d 639 (7th Cir. 1970). The issuance of a patent creates a presumption that it is worka *741 ble and any alleged unworkability of a patent does not preclude its relevance for determinations of obviousness. Blohm and Voss v. Prudential-Grace Lines, 489 F.2d 231 (4th Cir. 1973). Further, where a process has been fully disclosed in the prior art without a full appreciation of all its valuable attributes, the perception of new advantages in the old process does not constitute invention. Cutter Laboratories v. Lyophile-Cryochem Corporation, 179 F.2d 80 (9th Cir. 1949) . Accordingly, the prior 1937 and 1939 Finlayson patents will be given full effect for what they disclose even though there is no evidence that these process patents were ever placed in actual commercial use.

Defendants’ essential distinction of the Finlayson patents turns on the failure of the patents to disclose the 75 to 85° centigrade range of temperature. The ’973 patent is sustainable only if this range is a “critical” temperature range and if the range would not have been obvious to one skilled in the art who had knowledge of the prior art references. The court concludes that the ’973 patent fails on both counts.

One who seeks to establish a patent for a given range of conditions on the basis that he has located the optimum conditions for use of a given process faces a difficult task. It is generally not inventive or worthy of patent protection to point out the range of temperature at which the best results can be obtained. American Potato Dryers v. Peters, 184 F.2d 165 (4th Cir. 1950) . The location of optimum temperatures or conditions can usually be discovered by patient experimentation and does not constitute invention so as to entitle one to a patent monopoly. Helene Curtis Industries v. Sales Affiliates, 233 F.2d 148 (2nd Cir. 1956); National Lead Company v. Western Lead Products, 324 F.2d 539 (9th Cir. 1963); Triumph Hosiery Mills v. Alamance Industries, 299 F.2d 793 (4th Cir. 1962); Servo Corporation v. General Electric, 337 F.2d 716 (4th Cir. 1964). It is only where the selected point corresponds with a physical phenomenon and the patentee has discovered the point at which the physical phenomenon occurs that maintenance of a patent monopoly is permissible. Kwik Set, Inc. v. Welch Grape Juice, 86 F.2d 945 (2nd Cir. 1936).

The court rejects defendants’ position that they discovered the point at which the requisite physical phenomenon occurs, and finds nothing critical about the specified 75 to 85° centigrade range. At best, yarns steamed within the range produced marginally better commercial products; however, the properties of the yarn steamed inside and outside the specified range differed only in degree. Yarns steamed outside the specified range were approximately equal in commercial use and could be used interchangeably with yarns within the claimed range.

Even if the criticality of the claimed temperature is assumed, criticality alone will not suffice for invention. Servo Corporation v. General Electric, supra. It was well within the ability of one of ordinary skill in the art to have discovered the temperature range disclosed by the ’973 patent by the process of routine experimentation. Trial testimony by the experts for both parties is consistent that the Finlayson patents would teach one to post-treat cellulose yarns in a hot wet environment in the range of 90 to 100° centigrade. Defendants’ expert, Mr. Evans, testified on cross-examination as to how a man of ordinary skill in the art would have textured cellulose acetate before the ’973 patent. Mr. Evans testified that this hypothetical man would: (1) false twist under tension; (2) overfeed; (3) select the particular overfeed that would give him his desired effect; and (4) post-treat the yarn in a moist or wet hot environment. He further testified that in this post-heating treatment the prior art would have suggested working between 90 and 95° centigrade. Given that background, it is difficult to believe that a man of ordinary skill in the art would not eventually have reduced his temperature to 75 to 85° centigrade in order to obtain specifically desired yarn characteristics. In short, the “lengthy experiments” that led to the ’973 patent were the work of a skilled mechanic and not that of an inventor. See *742 Hotchkiss v. Greenwood, 11 How. 248, 266, 52 U.S. 248, 266, 13 L.Ed. 683, 691 (1851).

The court holds that the ’973 patent does not rise to the level of invention necessary to sustain patentability under 35 U.S.C. § 103 in that the claims asserted would have been obvious to one of ordinary skill in the art in light of the prior art disclosures, particularly the disclosures contained in Finlayson British Patent No. 464,981 and United States Patent No. 2,174,573.

Validity: Indefiniteness.

Plaintiffs have also attacked the ’973 patent for failure to comply with the requirements of 35 U.S.C. § 112, complaining that it does not contain a written description of the invention and of the manner of using it in such full, clear, concise and exact terminology as would enable a person skilled in the art to use or carry out the process.

Plaintiffs’ position on indefiniteness is that: (1) the patent is ambiguous as to its purpose in increasing the curliness of cellulose acetate yarn; (2) that the examples given within the patent are unworkable or are unclear and internally inconsistent; and (3) that the patent fails to disclose whether the wet bulb or dry bulb temperature is operable.

Plaintiffs’ first argument is rejected. The patent does indicate that the purpose of the invention was to improve the degree of curliness in cellulose acetate yarn by a second heat treatment. All parties have agreed that it was well known that the second heat treatment in fact reduces the curliness of false-twisted cellulose acetate yarn. The court concludes that this seeming ambiguity is explained in that the patentee was not comparing a pre-steamed false twist yarn with a post-steamed false twist yarn, but instead was comparing a false-twisted and steamed yarn with the yarn of the prior art products. Admittedly, there is some ambiguity on this point; however, the court is unconvinced the ambiguity is such that one of skill in the art would be confused or misled as to the purpose of the invention.

It was also established at trial that both examples contained in the ’973 patent contained inconsistencies. It was admitted that example 1 was unworkable and did not yield the desired result and that example 2 is unclear and is internally inconsistent. In fact, the admitted internal inconsistencies contained in example 2 were cited by defendants’ counsel as the reason that a formal disclaimer of claim 4, which called for a 6% shrinkage, was filed. Despite these difficulties, the court is unconvinced that the patent should fail for indefiniteness. The claims of the patent are straightforward in claiming a process requiring false-twisting, winding with considerable shrinkage, and subjecting yarns to a second heat treatment of 75 to 85° centigrade in the presence of steam while the yarn is substantially free from tension. The absence of a working example is not necessarily conclusive on the issue of indefiniteness. The test is whether there was sufficient working procedure for one skilled in the art to practice the claimed invention without undue experimentation. Application of Stephens, 529 F.2d 1343 (Oust. & Pat.App.1976). In light of the disclosures contained within the patent as a whole, the claims, and the relative skill of those in the art, the court finds that the absence of a working example does not render this patent invalid for indefiniteness.

The final assertion by plaintiffs is that the ’973 patent should be declared invalid for indefiniteness since it fails to disclose whether the wet bulb or dry bulb temperature is operative. In operating a steam box as specified in the ’973 patent, two temperature settings are generally utilized. The dry bulb temperature is the measure of the absolute temperature of the air in the steam box. The wet bulb temperature is the measure of the amount of moisture within the box and is normally used in conjunction with the dry bulb reading to ascertain the relative humidity of the moist atmosphere. Wet bulb temperature has also been described as the measure of the temperature which the yarn reaches while in a hot wet environment. In short, wet bulb is the measure of the temperature the *743 yarn reaches as compared to dry bulb which is the temperature that the air in the heater reaches. It has been admitted that both the dry bulb and wet bulb temperatures in the heater would affect yarn characteristics. At trial, defendants have asserted that it is only the wet bulb temperature that is operative in practicing the ’973 process and that those skilled in the art were aware that it was the wet bulb temperature which was referred to in the ’973 patent. For present purposes, the court deems it unnecessary to decide the “battle of the experts” as to whether wet bulb or dry bulb temperatures are operative. It is sufficient to note that those in the art were aware that there were two separate temperature settings involved in the operation of steam boxes. Given the 75 to 85° centigrade range, it would not have required undue experimentation by one practicing the ’973 process in order for a determination to be made of whether the wet bulb or dry bulb temperature yielded the desired yarn characteristics.

For purposes of the discussion on indefiniteness, the court has assumed that the ’973 patent is not void for obviousness in light of the prior art disclosures. Proceeding on this premise, the court finds that the disclosures within the ’973 patent were sufficient to enable one of ordinary skill in the art to understand the process claimed and to practice the claimed invention without having to resort to undue experimentation. United States Patent No. 3,123,973 is not invalid for indefiniteness.

Infringement.

For purposes of this discussion it is assumed that the ’973 patent is valid and is not unenforceable by reason of the prior findings of patent misuse. Plaintiffs have asserted three grounds of non-infringement: (1) that their sole purpose in overfeeding cellulose acetate yarn and subjecting it to a second heat treatment is to reduce (rather than to increase) the torque and liveliness of the yarn; (2) that the temperatures specified in the ’973 refer to dry bulb temperatures and many Throwsters have processed acetate outside the ranges specified; and (3) that cellulose acetate yarn textured by Throwsters’ specifications is not substantially free from tension and is not permitted to undergo considerable shrinkage.

Plaintiffs’ first argument has previously been rejected by the court in its discussion on indefiniteness. It is admitted by all parties that a post heat treatment reduces the degree of curliness in the yarn. This result is also disclosed in the patent (Column 2, Lines 36 and 37) where the effect of an additional thermal treatment is discussed and it is stated: “However, a little of the shrinkage capacity and of the springiness is lost.” Patentee’s references to increased curl and bulk were a result reached by a comparison of a yarn false twisted and steamed by the patented process with a yarn processed by other prior art techniques.

Plaintiff’s second argument is a partial non-infringement argument premised on this court’s making the determination that a dry bulb rather than a wet bulb temperature range is specified in the patent. It is not disputed that many Throwsters have steamed yarns at temperatures above 75 to 85° centigrade, dry bulb. The terms dry bulb and wet bulb have previously been defined in this court’s discussion of indefiniteness. In a “battle of the experts” at trial, defendants attempted to show the temperatures referred to in the ’973 were actually wet bulb temperatures while plaintiffs insisted that one skilled in the art would understand that the temperatures referred to the dry bulb temperature. The issue is close; however, the court is convinced by a preponderance of the evidence that one skilled in the art would understand that temperature references in the ’973 patent are to wet bulb temperatures. Sufficient evidence has been presented that the plaintiffs steamed acetate yarns at wet bulb temperatures within the range of 75 to 85° centigrade, and plaintiffs’ argument of non-infringement because of their use of dry bulb temperatures of above the patent range is rejected.

*744 Plaintiffs’ final argument on non-infringement is that the Throwsters have not processed yarns under conditions permitting the yarn to undergo “considerable shrinkage” and, accordingly, that Throwster processes do not come within the teachings of the ’973 patent. The court agrees and concludes that there is no infringement. The evidence establishes that from 1964 to 1970, plaintiff Throwsters processed acetate with typical specifications calling for between 10 and 16% overfeed to the take-up package. This rate of overfeed would give a net shrinkage in the range of 5 to 6% because of various rates of underfeed in the first heater. The argument is made that a 5 to 6% shrinkage cannot be deemed to be “considerable shrinkage.” In resolving this matter, the court has turned to the patent, the patent specifications, and the file wrapper. Throughout the prosecution of the ’973 patent prior art which disclosed low tension rewinding was distinguished from the patentee’s process since the prior art could not allow the type of “considerable shrinkage” that was an essential part of the process sought to be patented. The patent discloses that considerable shrinkage is to be achieved by overfeeding. Example 1 in the patent calls for a shrinkage of 33%. Claim 3 speaks in terms of shrinkage of from 30 to 35%. Example 2 in the patent is admittedly confusing but does call for overfeeds of about 20%. Further, the court notes that the ’973 patent was assigned to Chavanoz and that the Chavanoz processing specifications for cellulose acetate prior to 1969 called for package overfeeds of about 28%.

The procedures used by plaintiff Throwsters do not include package overfeeds within the range that the ’973 patent indicates could permit “considerable shrinkage.” The great majority of the plaintiffs’ specifications show overfeeds of from 8 to 14% and there is no evidence that any plaintiff has commercially employed package overfeeds as high as 20%.

The package overfeeds and resulting shrinkages employed by the plaintiffs are considerably less than the overfeeds and shrinkages suggested by the ’973 patent. Defendants have not discharged their burden of proof on infringement by establishing by a preponderance of the evidence that plaintiffs’ processing specifications for cellulosie yarns involved a crucial step required in defendants’ ’973 combination patent. The court is not able to conclude that plaintiffs’ processing specifications involve winding the yarn into a package under conditions whereby the yarn was permitted to undergo “considerable shrinkage” within the meaning of the ’973 patent. Accordingly, the court holds that the ’973 patent, even if valid, has not been infringed.

In addition to the facts found by the court in its treatment of the ’973 patent the court adopts as its own the following proposed findings of fact submitted by the parties on the patent issues:

1. On the issue of validity: plaintiffs’ proposed findings 9.2.1; 9.3.1 — 9.3.10 inclusive; 9.3.13-9.3.15 inclusive; 9.4.1; 9.5.1; 9.5.2; and 9.6.1.

Defendants’ proposed findings IX-30, 35, 37, 43, 44, 73-78 inclusive, and 84.

2. On the issue of infringement: plaintiffs’ proposed findings 9.1.1 — 9.1.4 inclusive; 9.1.6-9.1.17 inclusive; 9.7.1 with the exception of the last sentence; 9.7.2; and 9.7.5-9.7.9 inclusive.

Defendants’ proposed findings IX — 1, 2, 7, 9, 10, 11, 12 and 13.

Patent No. 3,165,881

Entitled “Production of High Bulk Yarns”, this invention, according to the specifications, “relates to improvements in the production of high bulk yarns.” The application for the patent was filed January 31, 1961, by the inventors, de Moncuit of Chavanoz and Crouzet of ARCT-France, with a claimed priority date of February 3, 1960. After surviving six rejections in the Patent Office over a period of four years six claims were finally allowed and the patent issued on January 19, 1965.

After reciting that synthetic yarns textured on conventional one heater false twist *745 machines have high bulk and high elastic elongation, the specifications state that while in some end products such as hosiery high elastic elongation is a desirable quality in a yarn, in others such as sweaters high bulk is desirable but “high elastic extension is disadvantageous.” Continuing, the specifications state:

“A further heat treatment of the high bulk yarn mitigates this disadvantage. The further heat treatment may be applied while stretching the yarn or the yarn may be further heated and then wound under low tension so as to permit a substantial contraction.

“The present invention is concerned with a machine which enables the whole operation of false twisting, heat setting and further heating to be carried out in a single run of the yarn and moreover enables the further heating to be effected either under stretching conditions or under low tension so as to permit contraction.

“The machine according to the present invention comprises a frame carrying for each yarn two heating elements and a false twisting spindle between the two heating elements, the false twisting spindle being constructed and mounted so that the false twist runs back into the heating zone of the first heating element and carrying feed devices in front of each heating element and behind the second heating element, the relative speeds of the three feed devices being adjustable.

* * * * * *

“The invention further comprises the combination of a voluminizing frame, as described above, with a creel. Such a creel may be of the magazine type, that is to say positions are provided for two [supply] packages for each yarn running through the voluminizing frame and are so disposed with respect to one another that the inner end of the yarn on one package . . . can be tied to the outer end of the yarn on the other package, the packages being of the type which are unwound by pulling over the end. Further, the frame of such a creel may be extended upwards and carry in its upper part take-up devices, one for each yarn, the frame being held at the top to the frame of the voluminizing frame by suitable crossbars. Most advantageously the combined machine is provided with a low platform between the voluminizing frame and the creel frame and the various thread guides are so mounted that the yarns pass from the [supply] packages under the platform, up through the elements of the voluminizing frame and back overhead to the take-up devices mounted in the upper part of the creel, thus enabling operatives on the platform to thread up or otherwise attend to the creel or to the voluminizing frame without interfering with the continuity of the process.”

In the diagram reproduced below there is shown one side of a two-sided machine with the supply and take-up packages located on the right-hand frame and the two heaters, false twist spindle and guide rolls located in the left-hand (“voluminizing”) frame. The two frames are joined by the platform at the bottom and a crossbar at the top.

In the diagram the yarn 5 travels from one of the supply packages 3 underneath the platform 12, upward through the guide *746 and feed rolls 16 and 17, through the first heater 18, through the false twist spindle 22, thence through the second heater 24, thence back across the passageway between the two frames and onto one of the take-up packages 31a.

Claim 1 of the patent reads:

“1. Yarn processing apparatus comprising a false twisting and yarn heating assembly for twisting and heat treating yarn, a yarn supply and take-up assembly disposed laterally to one side of said false twisting and heating assembly, said yarn supply and take-up assembly comprising yarn supply support means and yarn take-up means, and an intermediate operator supporting platform disposed between said assemblies, said platform having a supporting surface and a yarn passageway beneath said surface for passage of yarn between said assemblies while permitting an operator to stand on said platform without hindrance to the passage of yarn between said assemblies.”

Claim 2 is a dependent claim, calling for an apparatus according to claim 1 with the yarn supply and take-up means disposed one above the other. Claim 3 calls for a double-sided yarn voluminizing frame with creel frames on either side. Claim 4 adds descriptive language relating to yarn guides, crossbars connecting the overhead portions of the frame, and the cyclic path of the yarn running on the machine. Claim 5 is a dependent claim calling for an apparatus according to claim 4 with the take-up means above the supply means so that the yarn can be fed cyclically through the machine to the take-ups. 70

Plaintiffs argue the invalidity of the ’881 patent on two grounds: (1) the invention was on sale in this country more than one year prior to the date of the application for the patent in the United States and is thus within the “on sale” exception of 35 U.S.C. § 102(b); and (2) the patent is void for obviousness under 35 U.S.C. § 103. 71

The defendants vigorously contend to the contrary, arguing that the ’881 patent represented a radical departure from the prior art and another outstanding achievement by Chavanoz in the advancement of the art of false twist texturing.

The “On Sale” Defense.

Plaintiffs urge that the ’881 patent is invalid under 35 U.S.C. § 102(b) which provides in part that:

“A person shall be entitled to a patent unless the invention was ... on sale in this country, more than one year prior to the date of the application for the patent in the United States . .”

After a tortuous four-year history in the Patent Office the ’881 patent was issued on January 19,1965, pursuant to an application filed on January 30, 1961. The relevant date for Section 102(b) purposes, therefore, is January 30, 1960, one year prior to the United States filing date. It is settled that a single instance of a sale prior to the critical date suffices to invalidate the patent. Maibohm v. RCA Victor Company, 89 F.2d 317 (4th Cir. 1937). It is the single instance of “on sale” which is relied upon here by the plaintiffs.

The essential facts underlying this instance are not in great dispute. They are summarized in the numbered paragraphs below.

*747 1. The first ARCT machine to embody the double frame concept of the '881 patent was the FTF machine. On January 5,1960, in response to a December 29, 1959 letter inquiry from Joseph H. Hamilton, then president of Burlington Throwing Company, regarding the availability of a double heater machine, Robert F. Waters, then the Whitin Machine Works ARCT salesman, wrote to Mr. Hamilton as follows:

“You asked me about the Chavanoz or ARCT FFT (sic) machine in your December 29th letter, and whether we were in a position to furnish some information on this machine.

“I have received three photographs from Leo Soep when he was here recently, and prototypes of the machine have been running in Chavanoz’s throwing plant. I would prefer not to send the photographs because I expect to be in the south during the next couple of weeks and discuss the machine in detail with you at that time. I might add, however, that they are quoting a delivery of an approximate twelve weeks. Although I have not received any price information Leo Soep estimated that it might cost about 50% more than the FT machine, but this is strictly a guess.”

Mr. Waters then continued with a description of the FTF machine and its multiframe arrangement (PX 308).

2. At the trial Hamilton testified that his letter of December 29,1959, and Waters’ response of January 5, 1960, were in connection with his constant efforts to get machinery manufacturers to develop a machine to make set yarns in a continuous operation. He said that he understood from this correspondence that the FTF machine was not then ready, but would be available in the future (Tr. 15, p. 3072).

3. Waters testified at the trial that twelve weeks was the minimum lead time required for manufacture and delivery of ARCT false twist machines.

4. As promised in his January 5 letter, Waters visited Hamilton in his office “a couple of weeks” later. During that meeting Hamilton was shown pictures of an ARCT machine embodying a separate creel and platform arrangement. He made a sketch of the arrangement (now attached to PX 308) which shows the two frames separated at the bottom by the platform and joined at the top by the crossbar, one frame carrying the supply and take-up packages and the other the texturizing elements. Waters gave Hamilton additional information about the machine at that time.

5. In his testimony Hamilton said that he was “trying to zero in on the commercial application and commercial introduction of a machine of this type that would give us new products. I was very, very hot on moving in this direction from a manufacturing and products standpoint.” (Tr. 16, p. 3269.)

Hamilton further testified:

“Well, Mr. Waters, myself and Mr. Waters and Burlington had a very good working relationship and he was as anxious as we were to collaborate in the commercial development or exploration of these new yarn processing devices, including the double heater approach.” (Tr. 16, p. 3270.)

6. On March 14, 1960 — six weeks after the critical date under Section 102(b) — a textile publication carried a story and photograph of a new ARCT double heater machine. The lead sentence stated:

“NEW YORK. — A new machine to turn out a stabilized bulk type yarn via a continuous one-step false twist operation, as opposed to the usual two-stages, will be marketed in the United States by Whitin Machine Works, Whitinsville, Mass.” (PX 1292.)

7. On May 6, 1960, Waters wrote Armitage of DMRC about the FTF machine. He asked the following question: “How do you feel about our promoting the sale and licensing of this machine in our territory?” (PX 643.)

8. On June 6, 1960, Waters wrote Soep of Chavanoz concerning Whitin’s uncertainty about the legalities of licensing the FTF machine, stating:

*748 “I discussed verbally and by letter the subject of the FTF machine with Norman Armitage and now following the Atlantic City fair I should like to feel free to pursue promotion and sale of this new machine.” (PX 645.)

9. On April 17, 1961, Waters wrote a memorandum to Dalton of Whitin briefing him on the ARCT program. In the memorandum it was stated:

“You are planning to visit the knitting arts exhibition next week where we will be introducing a new machine identified as the type FTF (False Twist Fixed). I hope this will serve to acquaint you with our activities.” (DX 849.)

10. The FTF machine that was exhibited in April, 1961, at the knitting arts exhibition was the first ’881 machine to reach the United States. It was also the first ’881 machine that Waters had ever seen, and prior to 1961 Whitin had never had any agreement with ARCT-France to purchase a machine embodying the invention of the ’881 patent. Whitin received its first FTF machine in late April, 1961, and its first delivery of such machine was to Celanese after the knitting arts exhibition (Tr. 81, pp. 15807-11).

The cases decided under 35 U.S.C. § 102(b) apply two different tests to determine whether an invention is “on sale”, the “on hand” test and the “commercial exploitation” test. Under either test the parties agree that plaintiffs bear the burden of establishing the “on sale” defense by clear and convincing evidence.

The cases are not clearly instructive as to which of the tests is to be applied in this Circuit, see Nicholson v. Mullis Engineering and Manufacturing Company, 315 F.2d 532 (4th Cir. 1963), cert. denied, 375 U.S. 828, 84 S.Ct. 72, 11 L.Ed.2d 60 (1963), but even if this court is free to apply the commercial exploitation test, it has concluded that the plaintiffs have not carried their burden.

Under the “on hand” test a completed sale, with or without delivery, is not required to establish a bar to patentability within the meaning of Section 102(b). An offer to sell, made to a prospective purchaser after the experimental stage of development of the invention has passed, the invention reduced to practice, and a device manufactured in its perfected form, places the device “on sale” within the meaning of the statute. An article is “on sale” if it is on hand ready to be delivered to any purchaser whether or not the article has been sold. Nicholson v. Mullis Engineering, supra.

After reconsidering its prior holdings the Second Circuit in Timely Products Corporation v. Arron, 523 F.2d 288 (2nd Cir. 1975), abandoned its adherence to the “on hand” doctrine and adopted the commercial exploitation test. In that case, in which plaintiffs here place great reliance, the court held that an article is “on sale” if an application for a patent is filed more than one year after the solicitation of an order for a specific article to be produced later where the following requisites are present:

“(1) The complete invention claimed must have been embodied in or obvious in view of the thing offered for sale. (Citations omitted.) Complete readability of the claim on the thing offered is not required because whatever is published (or on sale) more than one year prior to the filing of a patent application becomes part of the prior art over which the claim must be patentable .

“(2) The invention must have been tested sufficiently to verify that it is operable and commercially marketable. This is simply another way of expressing the principle that an invention cannot be offered for sale until it is completed, which requires not merely its conception but its reduction to practice

“(3) Finally, the sale must be primarily for profit rather than for experimental purposes. (Citations omitted.)” 72

The text continues — open the document to read the rest.