GARRECHT, Circuit Judge.

The infringement of Claim 4 of United States Patent No. 1,844,142, issued to Arthur E.H. Barili for a stuffed pastry machine, on February 9, 1932, is the question involved in this appeal.

The court below found that Claim 4 had been infringed by Achille Bianchi and the Marlo Packing Corporation, which is hereinafter referred to as Marlo, and granted an injunction against further infringement by either defendant. Barili v. Bianchi, D.C., 72 F. Supp. 766. Bianchi is the manufacturer of the accused machine and Marlo is its user. Both have appealed.

Barili also has appealed, with respect to the lower court's failure to order an accounting for damages and its failure to award a reasonable attorney's fee, and also with respect to the lower court's having ordered the case to trial only one day after an order had been entered setting aside a prior judgment in this case in favor of Bianchi and Marlo.

Claim 4 reads as follows:

"In a ravioli machine, a pair of intergeared rollers, one roller having indented molds and provided with axial cutters between said molds, the other roller being made with annular peripheral cutters adapted to be positioned between the molds of the first named roller at the point of contact of the rollers, all the molds being spaced apart so as to provide a wide margin between the cutters and the molds, means for feeding sheets of flour paste to the rollers, and a detachable open bottom hopper for guiding stuffing to the paste sheets on the rollers."

The questions involved in the Bianchi and Marlo appeal are (1) the validity of Claim 4 and (2) infringement by the accused machine.

1. Some Threshold Questions

Bianchi and Marlo contend that Claim 4 "is fully anticipated by the prior art and is utterly invalid"; and that it "is also invalid because of its failure to read on the disclosures in the patent specification and drawings, or in other words, because the patent specification and drawings do not support the claim".

In their answer, Bianchi and Marlo set up seven patents as having anticipated Barili's invention. Of these, however, only two are referred to in their briefs — Holmes No. 518,454, dated April 17, 1894, and Evans No. 1,094,320, dated April 21, 1914. In addition, in their brief Bianchi and Marlo argue that Claim 4 is also "fully anticipated" by Oleri No. 1,479,925, dated January 8, 1924. Finally, they assert that Tommasini No. 1,236,998, dated August 14, 1917, and the "Raviolara pamphlet", said to have been obtained by Bianchi in Europe in 1919, "were introduced into the record at the trial to show that automatic ravioli manufacturing machines were in use long prior to the filing of the patent application which resulted in the patent in suit."

(a) The Burden of Proof

At the outset, it should be observed, that "the grant of letters patent is prima facie evidence that the patentee is the first inventor of the device described in the letters patent, and of its novelty." Cantrell v. Wallick, 117 U.S. 689, 695, 6 S.Ct. 970, 974, 29 L.Ed. 1017; 3 Walker on Patents (Deller's Edition) § 701, page 2009.

Before a patent can be declared invalid because of anticipation, its lack of novelty must be established beyond a reasonable doubt. Cantrell v. Wallick, supra, 117 U.S. at pages 695, 696, 6 S.Ct. 970; 1 Walker § 63, pages 300-303; American Bell Telephone Co. v. People's Telephone Co., C.C.N.Y., 22 F. 309, 313, affirmed, 126 U.S. 1, 572, 8 S.Ct. 778, 31 LEd. 863; Searchlight Horn Co. v. Victor Talking Machine Co., D.C.N.J., 261 F. 395, 401.

Particularly heavy is the attacker's burden when the validity of the patent has been sustained by court findings. General Motors Corporation v. Kesling, 8 Cir., 164 F.2d 824, 827, certiorari denied on March 15, 1948, 333 U.S. ___, 68 S.Ct. 732, and the many cases there cited.

(b) An Improvement Patent is Entitled to a Reasonably Liberal Construction

Barili asserts that his invention is of a "primary" or "pioneer" character. After carefully comparing his teaching with that of the prior art, however, we find ourselves unable to agree with that conclusion. Claim 4 sets out, at most, an improvement in the art of producing machines for the making of ravioli.

Nevertheless, even a patent that teaches merely an improvement in a familiar process merits a reasonably liberal construction. In 2 Walker § 247, page 1212, the doctrine is thus stated:

"Even though an invention be not a pioneer, if it marks a decided step in the art, it will be entitled to the benefit of the rule of equivalents, though not in so liberal a degree as if the invention were of a primary character. [Case cited] Thus, a patent which is an improvement on an old machine may be very meritorious and entitled to liberal treatment. [Case cited] * * * Where an invention undoubtedly marks a substantial advance in the art, the patent is to be given a reasonably liberal construction so as to secure to inventors the rewards to which they are entitled." 2. The Prior Art

We will briefly consider the earlier patents upon which Bianchi and Marlo rely, as anticipating Barili's device.

(a) The Holmes Patent

This machine, according to Holmes's specification, "is adapted for producing articles [of pastry or confectionery] of tubular form or articles consisting of a tubular envelope and a suitable filling". Continuing, Holmes says:

"My invention is also adapted to produce solid or porous cylindrical sticks. The articles are preferably produced in continuous lengths and may be subsequently cut into smaller lengths."

We find it difficult to spell out ravioli from tubes or sticks, whether solid or porous. In a word, Holmes's machine could not produce ravioli, nor did he claim that it could do so.

In Topliff v. Topliff and Another, 145 U.S. 156, 161, 12 S.Ct. 825, 828, 36 L.Ed. 658, the court said:

"It is not sufficient to constitute an anticipation that the device relied upon might, by modification, be made to accomplish the function performed by the patent in question, if it were not designed by its maker, nor adapted, nor actually used, for the performance of such functions."

Furthermore several claims of the Barili patent were rejected on Holmes. But Claim 4, the one here in question, survived.

In 3 Walker § 700, page 2010, it is said:

"The presumption of validity is strengthened by the circumstance that the alleged anticipating patent was considered by the Patent Office in connection with the application for the patent in suit."

(b) The Evans Patent

According to the specification, "This invention relates generally to the manufacture of `feeding' cakes for animals". Evans does not provide for any filling or stuffing to be pressed between layers of dough. And the filling is one of the distinguishing features of an authentic raviolo.

In other words, Evans, like Holmes, had in mind a machine that was not intended to turn out ravioli. And, also as in the case of Holmes, Evans's patent "was considered by the Patent Office in connection with the application for the patent in suit".

(c) The Oleri Patent

This utensil is a hand-operated device resembling a rolling pin. The machine has no automatic means for stuffing material between two sheets of dough. On the contrary, according to the specification, it commences to operate only after the two layers of dough with the stuffing inserted therebetween are "firstly laid flatly upon a table, after which the shaper and cutter is rolled over the same, thereby forming, sealing and cutting the individual raviolis [sic] in one operation." Emphasis supplied.]

(d) The Tommasini Patent

Conceding that he was not the first to devise a machine for the manufacture of ravioli, Barili himself was the first to direct the attention of the Patent Office to the Tommasini invention.

Tommasini's patent, however, does not call for two sheets of paste. The disclosure envisages only one sheet. The machine consists of devices for feeding this single sheet of paste, for feeding the stuffing upon it, for cutting the paste into disks with peripheral indentations, and for folding back each disk into a half-moon shape, with the edges closed so as to contain the stuffing.

In Tommasini's device, the feed and cylinder movements must be "intermittent" or "step by step". This is because "It is necessary that the cylinders * * * should not commence to carry the ravioli [sic] with them before the latter has received the stuffing and it is also necessary that the stuffing should be fed to the ravioli [sic] while it is still stationary". Each of the cylinders has to make a complete revolution in order to turn out a single raviolo between them.

(e) Mario Eccher's "Raviolara"

Counsel for Bianchi and Marlo offered in evidence as prior art "a print of a number of automatic ravioli machines which was published prior to the First World War".

The sheet contains cuts of sixteen machines designed for the manufacture of alimentary paste products. In addition to an introductory statement, there are explanatory "cutlines" under each picture. All the printed matter is entirely in the Italian, German, French and Spanish languages. Not a single word in English is to be found on the entire sheet.

The exhibit violates Rule 12 of this court, which reads as follows:

"Whenever any record transmitted to this court shall contain any document, paper, testimony, or other proceeding in a foreign language, and the record does not also contain a translation of such document, paper, testimony, or other proceeding, made under the authority of the inferior court, or admitted to be correct, the record shall not be printed; but the case shall be reported to this court by the clerk, and the court will thereupon remand it back to the inferior court, in order that a translation may be there supplied and inserted in the record."

Nevertheless, because we happen to be able to make our own translation of the circular, we will consider it in this appeal.

The sheet is undated. It purports to be issued by one Mario Eccher, of Rovereto, Trentino, Italy, who holds himself out as engaged in the "construction of automatic machines for alimentary pastes".

One of the cuts shown on the sheet is that of the "Raviolara", which is the main Eccher disclosure relied upon by Bianchi and Marlo as anticipating Barili.

Bianchi testified that he "picked up" the circular in Italy in 1919. Barili, on the other hand, stated on the witness stand that "a patent was rejected on this in Italy on account of a patent I had prior to this"; that "this machine here was made after mine"; that he corresponded with Eccher a few years before getting his Italian patent; that he sent Eccher some photographs of his machine, and that Eccher "was supposed to pay me a certain sum for the use of my Italian patent"; that, after a patent for the Raviolara machine had been rejected in Italy because of the Barili Italian patent, the Raviolara was patented in France, "a couple of years after" the Italian patent was issued to Barili; and that Barili's Italian patent was issued "one or two years before I got the United States patent".

The Raviolara cut in evidence is small and imperfectly shows the details of the machine. No letters patent are offered to support it, nor is there any patent claim, specification, or drawing. These lacks, together with the lack of date on the circular and the conflict in the testimony, render the Raviolara evidence wholly ineffective in removing the presumption created by the issuance of the patent in suit and the holding of the court below that that patent, "and particularly claim 4 thereof, is good and valid in law".

In our view, none of the foregoing patents or disclosures in the prior art anticipate or invalidate the patent in suit.

3. The Patent in Suit (a) The Mechanics of the Device

We turn now to Barili's patent, to inquire in what respects it discloses, when set off against the prior art, a residuum of novelty and invention.

Paste of the proper consistency is delivered to two sets of double rollers, from two batches at opposite ends of the machine. The paste passes between the double rollers, the space between each rollers having been "calibrated", or adjusted, so as to give each of the two sheets of paste the desired thickness.

The two sheets of paste are fed toward each other from the opposite calibrating rollers until they meet between two intergeared "form and cutting rollers", through which the two sheets pass and emerge as finished ravioli.

Centrally positioned about the form rollers is a hopper containing the stuffing. The hopper has no bottom, but is provided with curved edges that are substantially concentric with the form rollers. The paste sheets may properly be said to form the bottom of the hopper, and as they pass under it the stuffing is "urged along" with the paste into the space between the rollers. The latter are adjusted so close together that the two sheets of paste are very firmly pressed together and no more stuffing is allowed to crowd between the sheets than that which has lodged in the cavities of the molds in one of the rollers, to be explained hereafter.

The construction of the form and cutting rollers is one of the most important and distinctive features of Barili's invention. One roller is provided with six rows of molds, there being two molds in each row. The paste sheet which passes over that roller is forced into the molds, the space thereby created being filled by the same stuffing that depresses the paste into the molds. The two rollers are adjustably fixed so close together that the two sheets of paste are very firmly pressed together, and no more stuffing is able to crowd between the sheets than that which has lodged in the cavities of the molds.

The roller that contains the molds is also provided with "axial" or longitudinal cutters. The other roller has "annular", "peripheral", or circumferential cutters. When the rollers rotate, the axial cutters pass between the annular cutters. The combined action of these two types of cutters is to slit the two paste sheets lengthwise and crosswise, with the result that square-shaped ravioli are produced.

Claim 4 provides that the molds shall be "spaced apart so as to provide a wide margin between the cutters and the molds." This wide margin furnishes the ravioli with edges of sufficient width to permit their being tightly sealed by the pressure of the rollers, thus preventing the stuffing from leaking out.

From the foregoing, it will be seen that a dozen ravioli can be produced by each synchronous revolution of the two rollers. The device "may be driven from any suitable source of power".

(b) The Machine Is a "Combination" Invention

Barili's patent discloses a "combination" device. For example, molds were known to the prior art (Evans); so were rollers between which alimentary paste was to be pressed (Holmes, Evans, and Tommasini); so were cutters or "teeth" (Evans, Oleri, and Tommasini); and so were hoppers containing stuffing (Holmes and Tommasini). Barili's contribution to the art was to combine these and other elements in such a form as to turn out ravioli in a different manner and with greater rapidity.

Such advance involves more than mere technical skill of which, in the language of the books, any one learned in the art might be capable. Barili's contribution amounts to actual invention: not invention, perhaps, of the highest order, but an invention that is patentable.

For one thing, Barili's machine is superior in two respects over the devices invented by Holmes and Tommasini, his closest rivals. Unlike Holmes, Berili provides wide flanges or margins for his ravioli, so that they may be tightly sealed and thus prevent the stuffing from leaking. Unlike Tommasini, Barili offers an invention capable of turning out not one but twelve ravioli with every sychronous rotation of the two form and cutting rollers.

In Loom Co. v. Higgins, 105 U.S. 580, 591, 592, 26 L.Ed. 1177, the court said:

"At this point we are constrained to say that we cannot yield our assent to the argument, that the combination of the different parts or elements for attaining the object in view was so obvious as to merit no title to invention. Now that it has succeeded, it may seem very plain to any one that he could have done it as well. This is often the case with inventions of the greatest merit. It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention. It was certainly a new and useful result to make a loom produce fifty yards a day when it never before had produced more than forty; and we think that the combination of elements by which this was effected, even if those elements were separately known before, was invention sufficient to form the basis of a patent." (c) The Specification and the Drawings Support Claim 4

Bianchi and Marlo contend that Claim 4 is invalid because, although the cutters, according to the claim, are "adapted to be positioned between the molds * * * at the point of contact of the rollers," the patent drawings and the specification "clearly show and describe the rollers as being spaced apart". It is argued that Claim 4 is therefore invalid, since it "improperly describes the relationship of the elements of a claimed combination."

The gist of the contention lies in the meaning of the word "contact". Bianchi and Marlo apparently assume that the word necessarily implies an actual, physical "touching".

There are several reasons why this narrow construction should not be allowed to invalidate the claim.

In the first place, considerable latitude in semantics is permitted to an inventor. As was said in H.J. Wheeler Salvage Co. v. Rinelli Guardino, D.C.N.Y., 295 F. 717, 727, "a patentee has the right to use such words as to him best describe his intention, and they will be so construed as to effectuate that result."

Second, the specification and the claims of a patent are not to be construed with legalistic rigidity. Here, as elsewhere in the law, "the letter killeth, but the spirit giveth life". In Topliff v. Topliff and Another, supra, 145 U.S. at page 171, 12 S. Ct. at page 831, the court said:

"The specification and claims of a patent, particularly if the invention be at all complicated, constitute one of the most difficult legal instruments to draw with accuracy; and, in view of the fact that valuable inventions are often placed in the hands of inexperienced persons to prepare such specifications and claims, it is no matter of surprise that the latter frequently fail to describe with requisite certainty the exact invention of the patentee, and err either in claiming that which the patentee had not in fact invented, or in omitting some element which was a valuable or essential part of his actual invention. * * * The object of the patent law is to secure to inventors a monopoly of what they have actually invented or discovered, and it ought not to be defeated by a too strict and technical adherence to the letter of the statute, or by the application of artificial rules of interpretation."

Third, there is respectable authority for the proposition that the word "contact" does not necessarily presuppose an actual physical touching. The primary definition of the word in Webster's New International Dictionary, Second Edition, is as follows:

" 1. A union or junction of bodies; a touching or meeting; as a disease communicated by contact; contact with the air."

And "meet" is thus defined in Webster's:

" 2. To come, by accident or design, into contact or proximity with, or into the presence or company of, by approach from opposite or different directions; * * *." [Emphasis supplied]

Finally, even in a strict and technical sense, it may correctly be stated that the two rollers do come into contact with each other as they rotate. We agree agree with Barili's contention that the cutters on the rollers are "constituent parts" of such rollers, and that the cutters of one roller do actually touch the "peripheral surface" of the other roller. This contact, we may add, takes place when the rollers press, form, and cut the sheets of paste. Without this contact of the cutters of one roller with the peripheral surface of the other roller there could be no effective cutting or slitting of the ravioli into the desired squares, just as a knife in the hands of a housewife, cutting out dough into desired shapes, must of necessity strike the kneading board during the cutting operation.

Accordingly, we hold that the specification and the drawings of the patent support Claim 4.

4. Infringement (a) The Accused Machine

Bianchi and Marlo introduced in evidence an exhibit consisting of five large and clear photographs showing different views of the accused machine. Bianchi identified the photographs as those of a device made by him and sold to Marlo.

The description of this device appearing in the alleged infringers' own brief points up the striking resemblance of their machine to Barili's patented disclosure:

"Defendants' machine comprises two large rollers, one having a plurality of indented molds and the other carrying a number of spaced annular cutters and also a plurality of so-called axial cutters. The rollers are spaced one from the other and they are arranged to rotate in opposite directions. Two sheets of dough are conveyed between the rollers, and a stuffing material supplied from a hopper above is fed between the sheets. The stuffing material is pressed into the molds between the sheets as the rollers rotate, thereby forming pieces of stuffed pastry, such as ravioli.

"It is to be noted that unlike the structure defined by claim 4 [of the Barili patent] defendants' accused machine includes one roller having molds but no cutters, while the other roller carries both sets of annular and axial cutters."

From the foregoing, it will be seen that one of the principal distinguishing features claimed by Bianchi and Marlo for their machine is the fact that their device has all the cutters on one roller, while Barili's carries the axial cutters on one roller and the annular cutters on the other.

(b) The Doctrine of Interchangeability

The mere transfer of the axial cutters from the roller containing the molds to the other roller, containing the annular peripheral cutters, does not avert infringement. The means and the function of cutting in the two machines are identical. Exactly apposite to this sort of transposition is the language found in 3 Walker § 463, page 1699:

"Changing the relative positions or reversal of the parts of a machine or manufacture does not avert infringement, where the parts transposed perform the same respective functions after the change as before."

Bianchi himself testified that the position of the cutters was immaterial:

"It doesn't matter a bit. You can put them any way you want to." (c) The So-Called "File Wrapper Estoppel"

The other difference between the machines that is urged by Bianchi and Marlo as averting infringement, is the detail that in their device "the rollers are separated appreciably from one another," while Barili's Claim 4 is "limited to a structure wherein the rollers are in contact with one another".

We have already fully discussed the interpretation of the word "contact" in connection with the question of the validity of the patent in suit. Bianchi and Marlo further contend, however, that as the result of a so-called "file wrapper estoppel", Barili cannot now "ignore these limitations and contend that an accused structure not having those limitations infringes."

In view of this contention, a glance at the file history of Barili's patent now becomes necessary.

Claim 29 was submitted after several previous rejections of other claims. This claim set forth that one of the rollers had indented molds and axial cutters between the molds, and that the other roller had annular peripheral cutters positioned between the molds.

The examiner rejected Claim 29 on the ground that it was "vague", asking: "How can one roller have cutters positioned between molds which are on the other roller?" Barili then canceled Claim 29 and substituted Claim 31.

The examiner rejected Claim 31, for the same "vagueness" that he had urged against Claim 29, and suggested that:

"* * * if the expression adapted to be were added after the word `cutters' in line 4 and the expression at the point of contact of the rollers were added after the word `roller' in line 5, the vagueness would be eliminated and the claim would no doubt be allowable."

The applicant Barili amended the claim as suggested, and Claim 4 resulted.

It thus will be seen that the emendation was suggested by the examiner, not to restrict the claim to rollers that actually touch each other, but to rid the claim of a locution that is self-contradictory; namely, that one roller has cutters positioned between molds that are on the other roller.

Bianchi and Marlo argue that Barili, having inserted a limitation into his claim in order to secure its allowance, cannot now ignore that limitation in charging infringement. They admit, however, that Barili's rollers must be "in spaced relation since the cutters project therefrom, otherwise, the rollers could not rotate."

Is it reasonable to suppose that the examiner himself suggested an amendment to Claim 4 that would have made Barili's machine inoperable?

Is it not more rational to assume that both the examiner and the inventor himself, in the light of the patent drawings and the specification, intended that after the amendment the rollers should continue to be, as the specification and the drawings admittedly show they should be, "in spaced relation" or in close proximity to each other?

To adopt the narrow meaning of "contact" insisted upon by Bianchi and Marlo, would be to have Claim 4 teach absolute nonsense; for it would direct that a physical impossibility should be attempted. Because of the projecting cutters, to which Bianchi and Marlo themselves direct attention, it would be absolutely impossible for the cylindrical bodies of the two rollers to touch each other, even when at rest. A fortiori, it would be impossible for them to rotate.

In view of the entire file wrapper history and in view of the policy of the law to protect a patented article from infringement, we cannot say that Barili's adoption of the examiner's suggestion as to "contact" deprives him of the right to complain of infringement by Bianchi's machine.

(d) Infringement Is a Matter of Fact and of Substance

A leading text-writer on the law of patents has pointed out that infringement is a question of fact. 3 Walker § 450, page 1680.

It is also a question of substance, and not of nomenclature. It is not to be settled by striving to ascertain the difference between tweedledum and tweedledee.

In Hydraulic Press Mfg. Co. v. Williams, White Co., 7 Cir., 165 F.2d 489, 492, the court said:

"In determining the question of infringement, the court is not to judge about similarities or differences by the names of things, but is to look at the machines or their several devices or elements in the light of what they do, or what office or function they perform, and how they perform it. [Case cited] * * * One does not escape infringement by providing a single element which fully responds to a plurality of elements in the patent. [Case cited]"

So here, Bianchi did not escape infringement by putting all his cutters on one roller, for he thereby was "providing a single element which fully responds to a plurality of elements [i.e., two cutting rollers] in the patent". Indeed, as we have already noticed, Bianchi himself conceded that the rollers can be put "any way you want to."

Nor need the substantial indentity between the two machines be demonstrated to a mathematical certainty. In City of Grafton, W.V. v. Otis Elevator Co., 4 Cir., 166 F.2d 816, 821, the following language was used:

"Rarely do we find an example of what might be called perfect infringement. No patent infringer would be so silly as to make and vend a device similar in every minute detail to a patent. Infringement connotes, between the patent and the accused device, merely correspondence as to the substantial, dominate and essential elements. Any other view would make of a patent a foolish and fatuous thing."

After a careful examination of the record and of the exhibits, we cannot say that the court below erred in its factual finding that the accused machine, manufactured by Bianchi and used by Marlo, infringes Claim 4 of Barili's patent.

(e) The Statute of Limitations

While the court below found that there was infringement, it held that Barili had "failed to produce evidence of any alleged damage." In view of Barili's appeal from the part of the judgment denying damages, it becomes necessary for us to inquire whether the record supports this sweeping negative holding of the trial court. Since 35 U.S.C.A. § 70 provides that "recovery shall not be had for any infringement committed more than six years prior to the filing of the complaint in the action", we will limit our inquiry to the six-year period preceding October 19, 1946.

Barili testified that in the summer of 1946 Herbert Gierth, Marlo's superintendent, had said to him that Marlo had had a machine made by Bianchi "just exactly like" Barili's "and twice as wide", but that they had not used it for "five or six months".

Gierth himself at first denied that he had made any such statement at the time in question, but later in his testimony made statements from which partial corroboration of Barili might be spelled out:

"Q. Did you tell Mr. Barili that you had bought a machine from Bianchi that was just like Mr. Barili's machine, only twice as wide? A. No, not at that time. There was no occasion of saying that. In fact, I didn't know this.

"Q. Did you state that you had bought one machine four or five years ago like Mr. Barili's machine, only twice as wide? A. I told him we bought this ravioli machine from Mr. Bianchi shortly before the war broke out. This machine has been in use about half a year prior to the outbreak of the war, and then we discontinued it, because we are going into Army and Navy contracts, and has not been used since that time, and still is not in operation today."

Gierth described the machine to which he was referring as having one roll with the molds and the other with "the cutters, cross and lengthwise". That description fits the accused machine. "Half a year prior to the outbreak of the war" would be about June, 1941, well within the six-year statutory period.

Bianchi himself testified that within six years prior to the time that he was speaking, he had "made some rollers with the pocket [mold], the margin, and the cutter all in one roller, which we have a patent". A moment later, however, he insisted that "the last six years I didn't make any".

Barili called as one of his witnesses Carlo Cortopassi, a machinist at the Bianchi Machine Shop, and in his reply brief he relies upon fragments of Cortopassi's testimony as supporting the contention that "the machine" for Marlo was built in Bianchi's shop six or seven months before the war. But when it comes to identifying "the machine" as the accused device, Cortopassi's testimony is wholly ineffective. Indeed, probably because of language difficulties, some of Cortopassi's statements on the stand are unintelligible.

In any event, at the close of his direct testimony Cortopassi admitted that he had made only "part of the rollers", had never seen the machine in operation, had never seen a raviolo come out of it, and had never seen any cutters on the rollers.

And even as to the time element, the witness was extremely indefinite. In a portion of his testimony not completely quoted in Barili's brief, Cortopassi gave the date that "the machine" was made in Bianchi's shop as follows:

"Before the war, I guess, six or seven months — six or seven years ago. More than that".

The italicized words are omitted in Barili's quotation.

From the foregoing, it will be seen that the testimony on the subject of infringement within the six-year period is both vague and conflicting. While we realize that the burden of proving that the infringement did not occur within the six-year term lies with the defendants Bianchi and Marlo (Peters v. Hanger, 4 Cir., 134 F. 586, 590), we cannot say that the trial judge, who heard and saw all the witnesses, erred in finding that Barili had failed to produce evidence of damages. It must be remembered that all the evidence regarding infringement within the statutory period was oral.

Since the court found that there were no damages, it was correct in not ordering an accounting. See O'Cedar Corporation v. F.W. Woolworth Co., 7 Cir., 73 F.2d 366, 367, quoted by Barili himself.

Barili also complains that the lower court ordered the case to trial the day after setting aside a former judgment in favor of Bianchi and Marlo. He also asserts that the court erred in not awarding him a reasonable attorney's fee. Neither of these points requires discussion, since both refer to matters within the discretion of the trial court, and our examination of the record convinces us that the court did not abuse its discretion.

Accordingly, the judgment is affirmed in its entirety.