HAMLEY, Circuit Judge.

Amerio Contact Plate Freezers, Inc. (Amerio), brought this action to have Knowles United States Patent No. 2,-927,443 declared invalid and not infringed by Amerio. As assignee of competing MacKenzie United States Patent No. 2,697,920, the company also sought an adjudication, under 35 U.S.C. § 291, concerning interfering patents. Both patents are for machines which automatically freeze packaged foods. By denial and counterclaim the defendants, Frank W. Knowles, applicant for the Knowles patent, and Belt-Ice Corporation (Belt-Ice), assignee of that patent, joined issue on the allegations of the complaint.

After a trial to the court without a jury, judgment was rendered for defendants. Plaintiff appeals specifying as error findings of fact and conclusions of law bearing on all three issues — validity of the Knowles patent, its infringement, and interference between the Knowles and MacKenzie patents. In its briefs on appeal, however, appellant has not dealt with infringement as a separate issue.

Appellant contends that the trial court should have held the Knowles patent invalid on the ground that the Knowles invention was on sale more than one year prior to the date of the application for that patent. Reliance is thereby placed on 35 U.S.C. § 102(b) which provides, in part, that a patent may not be granted if the invention has been in public use or on sale in this country more than one year prior to the date of the application for patent in the United States.

The application for the Knowles patent was filed on May 23, 1952, so that the critical date insofar as section 102(b) is concerned, is May 23, 1951. The trial court premised its conclusion of law that the Knowles patent was not on sale prior to May 23, 1951, on two crucial findings of fact, each of which is here challenged by appellant. One of these findings, and that to which we will first direct attention, is to the effect that appellees had not constructed a machine embodying the Knowles invention prior to May 23, 1951. 1

In September and October, 1950, Knowles had workmen make a mock-up of a machine embodying the essential *461 features of the Knowles invention. 2 It contained moving parts and could be operated to demonstrate how a freezer machine incorporating the invention would work.

But the mock-up was only a model, built on a reduced scale, and made from crating lumber and plywood. It was not capable of freezing packaged foods. Its construction was accordingly not that of a machine embodying the Knowles invention.

Prior to May 23, 1951, Knowles had under construction two machines embodying the invention which were intended to be fully operating freezer machines. At least one of these machines was near completion on the critical date.

Knowles began ordering materials for use in constructing these machines as early as January 2, 1951. The work of preparing plans began on the following day. On January 23, 1951, he ordered thirty refrigerator plates from Dole Refrigerating Company for installation in the machines. 3 In March or April, 1951, appellees arranged to have the two machines constructed in the Carmac Shipyards in Seattle.

The date set for completion was July 5, 1951; as Knowles testified, this was when the machines were required by Evergreen Frozen Foods Co. (Evergreen) for the pea season. Knowles testified that the major portion of the work on the machines was done before May 2, and that one machine was “definitely assembled” by May 15, except for the installation of the freezer plates. Installation of the plates required that they be dropped into place after which steel reinforcing bars and other minor parts were added.

Knowles testified that the plates did not arrive until May 22. 4 Before the plates arrived, all testing which could be done without the presence of the plates had been completed and was successful. The plate control mechanism, however, could not be tested until the plates were installed. 5

When the critical date of May 23,1951 arrived, however, there was still considerable work to be done before either of these machines would be fully operative. Work on the plates after their arrival was in progress at the Belt-Ice plant on May 23. On May 25, Knowles took two of them over to the Carmac yards and for the first time placed plates in one of these machines. The placing of plates proceeded with«ut interruption after this initial installation, but the work of adding the reinforcing bars and other minor parts was in progress until June 16. 6

*462 After the plates had been received and installed the machines were given operating tests in the Carmac shipyards. However, it was necessary to postpone final testing until the machines could be set up at the Evergreen plant because there was no refrigerant or compression system at the shipyards.

The machines were shipped to the Evergreen plant at Snohomish on June 19 and 21, 1951. They were then installed, hooked up to a refrigeration system, operated and tested as fulfy completed freezers.

This testing was conducted by Belt-Ice employees and consisted of the introduction of ammonia vapor under pressure to determine if all the joints were right, and all the hoses were tight. The machines operated satisfactorily except for the development of some improper icing conditions which were corrected by the application of antifreeze to critical parts.

Testing by Belt-Ice employees was in progress until July 9, and on July 12 Knowles wrote in his diary: “Job finished. We have a freezer.” The two machines were used to freeze peas, corn, strawberries and raspberries during the months of July, August and September, 1951.

The evidence reviewed above, which is not disputed, amply demonstrates that neither of the machines being constructed for Evergreen was completed prior to May 23, 1951. It is not asserted that any other machines embodying the invention were completed, or even under construction, prior to that date.

In our opinion the finding of fact that appellees did not, prior to May 23, 1951, construct a machine embodying the Knowles invention, is not clearly erroneous. 7

The second finding of fact questioned on this appeal is to the effect that there was no definite offer made by one party to buy or sell a machine embodying the Knowles invention, and accepted by the other, prior to May 23, 1951. Appellant argues that a contract between Knowles and Evergreen for the construction, sale and purchase of the two machines referred to above was in existence prior to that date.

Drawings pertaining to the invention, and the mock-up referred to above, were shown to potential customers late in 1950. They were shown to Cedargreen, manager of Evergreen, on December 20, 1950. At that time Cedargreen indicated an interest in obtaining two freezer-machines for Evergreen. 8 On January 2, 1951, Knowles saw Cedargreen again and it was on this date that Knowles began ordering materials needed for the construction of two such machines. On-January 9, Knowles visited the Evergreen plant in Snohomish, Washington, and saw Cedargreen at that time.

On January 30, 1951, Belt-Ice quoted’ to Evergreen two fifteen-station Knowles.. *463 automatic package freezers at $11,400, f. o. b. Seattle. The terms stated were twenty-five percent with the order, fifty percent upon delivery, and twenty-five percent after thirty days of operation. Knowles testified, in effect, that he knew at this time that Evergreen would buy two of the freezers “if they could finance them.”

On February 14, 1951, Cedargreen, replying to the letter of January 30, 1951, stated that Evergreen could not purchase the freezers outright. Evergreen indicated, however, that it might be able to buy them on a rental basis, paying a cent a pound upon products frozen in the machine, such rental to apply on the purchase price. The next day Knowles replied that he was “willing to go along on a deal of that sort if we could have an advance to aid in building the freezers.”

In a deposition which Knowles had given in January, 1958, in Patent Office Interference No. 88,174, Knowles made statements to the effect that these machines were being constructed pursuant to an order from Evergreen received early in 1951. 9

At the 1962 trial, however, Knowles testified that he had no order for these machines until June, 1951, and that they were built on “speculation.”

The assignment of one of the job orders to those freezers states that they were “for stock.” On June 5, 1951, Knowles noted in his diary: “R.O.C. came in. Looked at freezers. Agreed to write us a letter outlining payment of one cent per pound up to 5,000 [sic. 500,-000] pounds, then one-half cent.” A purchase order to this effect was written on June 23, 1951. 10

The court apparently considered that Knowles’ trial court testimony as to when the contract with Evergreen had been entered into was more accurate than his inconsistent testimony in the 1958 deposition. The written communications between the parties and Knowles’ diary entries support this view. According to these items of evidence no oral contract was entered into prior to June 5, 1951, and no written contract until June 23.

We conclude that the finding that no contract was entered into between these parties prior to May 23, 1951, is not clearly erroneous.

Apart from its attack upon the findings of fact discussed above, appellant argues that the trial court erred in concluding that the Knowles invention was not on sale, within the meaning of section 102(b), prior to May 23, 1951.

The correctness of this ruling is to be judged in the light of the entire course of conduct of Knowles and Belt-Ice. This will include the now-established facts that no fully-operative machine embodying the Knowles invention had been completed, and no contract for the construction and sale of such a machine had been entered into, prior to the critical date.

The latter facts are sufficient to distinguish every appellate decision, involving an invention incorporated in an ar *464 tide for sale, upon which appellant relies in attacking this conclusion of law. Those are all cases where, prior to the crucial date, there was an offer or a contract to sell an article or apparatus embodying the invention, and at least one fully-operative article or apparatus of that kind was completed prior to the critical date. 11

Where no fully-operative article or apparatus incorporating the invention was in existence prior to the critical date, the appellate courts have uniformly held that a contract, prior to that date, to sell such an article or apparatus, is not a placing of the invention on sale in the sense intended by the statute. 12 If a contract to sell a machine embodying the invention does not place the invention on sale prior to the time a fully-operative machine comes into existence, it would seem that an offer to enter into such a contract could have no greater efficacy in producing “on sale” conduct. It follows that, if we are to adhere to the decisional law last cited, the trial court conclusion was correct whether a contract had been entered into, or only an offer had been made, prior to May 23, 1951.

Appellant calls attention to a number of district court decisions which, in its opinion, stand for the proposition that an article or apparatus embodying the invention may be on sale prior to the crucial date, within the meaning of section 102(b), even though no fully-operative article or apparatus of that kind is in existence prior to that date.

In view of the available appellate court decisional law bearing on the point, as indicated above, we do not believe that an examination of all of these district court decisions (which, appellees assert can for the most part be distinguished) would be helpful. However, appellant contends, in effect, that the citation of Philco Corporation v. Admiral Corporation, D.C.Del., 199 F.Supp. 797, and Chicopee Mfg. Corp. v. Columbus Fiber Mills Co., D.C.Ga., 165 F.Supp. 307, in our recent Tucker decision (note 11), represents acceptance by this court of all that Philco and Chicopee stand for.

We do not regard Chicopee as holding that an offer to sell an article embodying-the -invention places the invention on sale prior to the time that a completed article embodying the invention is in existence. In that case samples of the cloth embodying the invention were in existence and the offer was made with reference to the samples. It seems to us that a sample of a cloth embodying an invention is a fully-completed article of the kind proposed to be sold. The mock-up of the Knowles machine was not a sample-but only a non-operative model, and hence not a completed article embodying the invention.

In Philco, the district court held that a showing of non-operative mock-ups of a television set embodying the invention, for the purpose of soliciting orders, was a sufficient competitive exploitation to constitute a placing of the invention on-sale. At the time of such showing there were in existence several completely operative but handmade sets embodying the invention. These sets were then undergoing final tests and no production runs had yet been made.

Prior to the critical date and as a result of such showing, the prospective buyer, Firestone Tire and Rubber Co., executed certain documents entitled “releases” which, under the long-standing business relationship between the two-companies, Philco considered as equivalent to a purchase for its purposes. The “releases” were not, however, binding purchase orders in the normal sense. In holding that the Philco activity *465 amounted to a placing of the invention “on sale,” the district court used language which indicates that it would probably have reached the same result if the handmade sets had not been in existence prior to the crucial date.

But while Philco, (but not Chicopee) may thus be taken as a district court decision contrary to the conclusion we have expressed above, our reference thereto in Tucker did not constitute an acceptance of that position. In Tucker, articles embodying the invention were concededly in existence prior to the critical date and we there had no occasion to consider what our view would be if this were not the case.

In Tucker both Chicopee and Philco were cited only on the question of whether it was necessary that a contract to sell articles embodying the invention, entered into prior to the critical date, must be performed to the extent of delivery prior to that date, in order for there to have been “on sale” activity before that date. We held that delivery was not necessary. In the case now before us there was no completed article capable of being delivered before the critical date.

The purpose of the “on sale” provision of the statute is to prevent an inventor from having the advantage of a monopoly for more than the statutory period by engaging in the competitive exploitation of his invention after it is fully developed before applying for a patent. Cf. Metallizing Engineering Co. v. Kenyon Bearing & Auto Parts Co., 2 Cir., 153 F.2d 516, 518. 13

Ordinarily, however, selling activity (at least activity short of actual sales) prior to the time that a fully-operative article or apparatus incorporating the invention comes into existence, is not a reliable indicium of competitive exploitation. Until at least an operative prototype has been completed and tested, the competitive effectiveness of such activity, in all probability, will be impaired by the aura of continuing developmental, experimental and testing effort. Moreover, at this stage, such activity is likely to be more for the purpose of eliciting needed changes in design and testing whether the market potential warrants continuance of the project, than to launch full-fledged commercial exploitation.

We need not decide whether a different result would have been called for in our case if a contract to buy and sell the machines in question had been entered into before May 23, 1951. 14 Nor do we hold that circumstances could not be stated which would warrant a conclusion that an invention was on sale prior to the crucial date even though no prototype was then in existence, and no binding contract had been entered into. 15

Here there was not a contract prior to May 23, 1951. Nor were there special circumstances which required the court to hold, despite the absence of a contract and of a fully-completed prototype, that the invention was on sale, in the statutory sense, prior to that date.

In our opinion the trial court did not err in concluding that the Knowles invention was not on sale prior to May 23, 1951. We accordingly sustain the holding of that court, as against the section 102(b) contention, that the Knowles invention is valid and has been infringed by appellant.

The remaining question has to do with the issue raised in appellant’s second cause of action concerning alleged interference between the Knowles and Mac *466 Kenzie patents. In this connection appellant, predicating district court jurisdiction on 35 U.S.C. § 291, invoked the second pragraph of 35 U.S.C. § 135, now 35 U.S.C. § 135(b). Under this provision one may not, in any patent application, make a claim which is the same as, or for the same or substantially the same subject matter as, a claim of an issued patent, unless such a claim is made prior to one year from the date on which the patent was granted.

On December 28, 1954, while the Knowles patent application was pending in the United States Patent Office, that office issued MacKenzie Patent No. 2,-697,920. In Patent Office Interference No. 88,174 between the MacKenzie patent and the Knowles patent application it was contended on behalf of the MacKenzie patent that Knowles was advancing claims which were untimely under section 135. In support of this view it was asserted that the first time Knowles advanced claims which conformed closely to MacKenzie Patent claims 3, 6, 7, 10 and 12, was on March 23, 1956, when Knowles added claims 25, 26, 27, 28 and 29 to his pending application.

During the motion period in the patent office proceedings the Primary Examiner rejected a like contention. He ruled, in effect, that the code is satisfied if an applicant, in any of his claims, had claimed the subject matter of his invention within the critical year, and that Knowles had done so with regard to the claims he filed on March 23, 1956. When the interference proceeding came before the Patent Office Board of Patent Interferences, the same conclusion was reached, and Knowles was awarded priority with regard to his March 23, 1956 claims. Patent Office Interference No. 88,174. 16

Appellant then began a district court proceeding to review that decision of the Patent Office Board. Commenced in the same district court as that in which the suit now before us was filed, it was docketed as Civil Action No. 5092, and consolidated for trial with the instant suit (Civil Action No. 5171). As before indicated, appellant made the interference question the subject of the second cause of action in the suit here on appeal.

In No. 5092 judgment was entered for Knowles and Belt-Ice, and against Amerio, thus sustaining the decision of the Patent Office Board. Amerio has not appealed from that decision but has specified as error the adverse ruling of the district court on its second cause of action in the suit before us, in which the interference question was independently raised.

Appellees argue, and we agree, that the adjudication in No. 5092 is res judicata on this question. Were we to here consider the interference question on the merits, and reverse, there would be in existence two inconsistent final judgments. One would sustain the Patent Office Board determination awarding Knowles priority on the questioned claims, and the other would, in effect, award priority to MacKenzie. Appellant has not asked us to regard the instant appeal as also bringing on for review the judgment in No. 5092, and in any event we find no warrant for doing so.

The judgment is affirmed.

Notes

. Appellant specified as error tlie finding “ * * * that appellant (sic — appellees) did not construct a machine containing the invention prior to * * * ” May 23, 1951. This specification makes reference to paragraphs 20 and 21 of the findings, but those paragraphs contain no recital to that effect. In paragraph 26, however, such a finding is in effect made. While the specifications of error make no express reference to paragraph 28, we will regard it as included in the specification referred to above, in view of the similarity of subject matter.

. The Knowles invention consists primarily of a cage containing a movable stack of refrigerated plates which may be operated so as to automatically freeze packaged foods. The plates are moved by means of an elevator in such a way that successive pairs of plates can be brought level with a loading port and opened for the insertion of packages. The plates are held in place for this operation by control elements called “dogs.” A machine incorporating the invention loads from the front and unloads from the rear at a single level.

. In this invention the element which makes products freeze is an ammonia refrigerant which is circulated through refrigeration contact plates. Without these plates the machine could not perform its function.

. Knowles’ diary contains a May 15, 1951 entry: “Plates arrived.” But on May 18, 1951, be wrote in bis diary: “No plates as yet.”

. Knowles testified that be knew the plates would work because they bad been in use for years. But, until the plates were completely installed and the machine set up, it was not possible to test, by mechanical means, whether the plates would stop at the right location, engage the dogs properly, and move all the way up and down.

. Four steel reinforcing bars were needed for each plate. These bars were made at the Belt-Ice plant. After the plates were dropped into place, each bar had to be drilled, tapped and screwed to the plate. With regard to this, Knowles testified : “ * * * it is a lot of work, and that wouldn’t be done until the plates arrived.” Counterweight rods, constructed and installed after placing- the plates, *462 were also required so that tlie machines could accommodate heavy loads.

The following entries were made in Knowles’ diary, covering the period from May 28 to June 16,1951:

May 28 — “ * * * Package freezer going together okeh.”

June 1 — “ * * * Motor for package freezers shipped.”

June 4 — “Still working on package freezers. * * * ”

June 7 — “Freezers coming along * * * Motors Monday.”

June 8 — “Freezers still coming along. # ip ijs >)

June 16 — “Shop and Carmac * * * Package freezers about finished. R. O.C. [R. O. Cedargreen, manager of Evergreen] will pick up first one Monday a.m. We will live with it from then on.”

. Since appellant is asserting the “on sale” defense, it had the burden of proof to establish the facts which would call that defense into play. In order to sustain this burden appellant was not required to prove these facts beyond a reasonable doubt. It was, however, required to prove them by substantial, clear and satisfactory evidence. Tucker Aluminum Products, Inc. v. Grossman, 9 Cir., 312 F.2A 293, decided January 10, 1963.

. On that date Knowles made this entry in his diary: “R.O.C. [Cedargreen] in wants a couple of freezers. Guess we are going to have to build them.”

. In this deposition the following colloquy, occurred:

“Q 74. What was the next activity regarding the package freezer following the construction of this prototype? A. I actually sold one, — I sold two, rather, to the Evergreen Frozen Foods from the prototype and the preliminary sketches. Q 75. What do you mean by sold? A. Well, they said they would take two of them. Q 76. Who are ‘they’? A. R. O. Cedargreen, was the manager of the place.”

“Q 344. Why was Belt-Ice anxious to obtain these plates, Mr. Knowles? A. I had taken an order for two package freezers into wliicb these plates were to go, from the Evergreen Frozen Foods, and had verbally agreed that delivery would be made the first of June, so that the plant would be ready for peas about the first of July.”

. The terms of payment stated in this letter were one cent per pound for all produce frozen up to 500,000 pounds and one-half cent per pound thereafter throughout the 1951 freezing season; the balance thereafter, if any, to bear six percent interest from the end of the season until paid in full; the same terms to apply to the 1952 season.

. Contract to sell: Tucker Aluminum Products, Inc. v. Grossman, 9 Cir., 312 F.2d 293; offer to sell; Wende v. Horine, 7 Cir., 225 F. 501; cf. Magee v. Coca-Cola Company, 7 Cir., 232 F.2d 596.

. B. F. Sturtevant Co. v. Massachusetts Hair & Felt Co., 1 Cir., 124 F.2d 95; Burke Electric Company v. Independent Pneumatic Tool Co., 2 Cir., 232 F. 145, 234 F. 93; McCreery Engineering Co. v. Massachusetts Fan Co., 1 Cir., 195 F. 498.

. Congress originally set the permissible period of “on sale” activity prior to patent application at two years. Section 7, Act March 3, 1839, 5 Stat. 353, 354. Demonstrating its continued adherence, and perhaps increased interest, in that policy, Congress, in 1939, shortened this permissible period to one year. Section 1, Act of August 5, 1939, 53 Stat. 1212.

. But see the cases cited in note 12.

. Philco may have been such a case. The transaction there in question, involving companies which had long dealt with each other on a special basis, may well have given the seller a competitive advantage that could properly be equated with “on sale” activity.

. The Board said, in its decision:

“On the second question in issue, that of estoppel under 35 U.S.C. 135 on 'whether or not Knowles had claimed substantially the same invention at some time up to within a year of the grant of the MacKenzie patent we consider that the Primary Examiner arrived at the correct conclusion in holding Knowles not to be estopped. * * * As explained in Rieser v. Williams, 733 O.G. 572, 118 U.S.P.Q. 96 every limitation in the count claim of the patent is material as regards right to make or priority, but is not material as regards claiming by the applicant before the lapse of the statutory year unless there is a difference in the limitation that is inventive or critical to patentability. * * * The party Knowles accordingly is held not barred by estoppel under 35 U.S.C. 135 * *