On July 24, 1959, this action was commenced by Paper Converting Machine Co., Inc., alleging that Hudson-Sharp Machine Company was infringing United States Letters Patent No. 2,870,840, issued to Edwin M. Kwitek, vice president of Paper Converting Machine Co., Inc., to whom the patent was later assigned. Application for the patent, entitled 'Web Cutting Apparatus,' was made May 16, 1957, and the patent was issued January 27, 1959.

The defendant answered the complaint, denied infringement, alleged invalidity and invoked a defense of file wrapper estoppel. The original defendant was merged with Food Machinery and Chemical Corporation (FMC), the instant defendant.

The case was tried by Judge Robert E. Tehan, and the proceedings extended over a period of some six years. During that time numerous discoveries were sought and obtained by the respective parties, interrogatories were proposed and answered, four pre-trial conferences were held, a three-week trial was had with a full day devoted to oral argument, and four post-trial briefs were submitted.

The court, on June 5, 1967, rendered its opinion, in which it decided the issue of infringement in favor of plaintiff and rejected defendant's defense of file wrapper estoppel. Paper Converting Machine Co., Inc. v. FMC Corporation, 274 F. Supp. 372.1 Consistent with the opinion, findings of fact and conclusions of law were prepared by counsel for plaintiff, submitted to opposing counsel as to form and, with slight changes, adopted by the court. The court's findings are forty-eight in number and cover twenty pages of the appendix. The court concluded that defendant's perforators infringed claims 1,2 2, 3, 4, 8 and 9 (those involved in the suit):

'2. * * * because the machines embody the construction defined by each claim and operate in the same fashion to produce the same advantages. In determining whether a machine infringes, similarities or differences are to be determined, not by the name of things, but in the light of what they do. Union Paper Bag Machinery Co. v. Murphy, 97 U.S. 120, 125, 24 L. Ed. 935 (1897); Nordberg Mfg. Co. v. Woolery Machine Co., 79 F.2d 685, 692 (7th Cir. 1935).

'3. Where the accused machine falls clearly within the claim, infringement is made out, and that is the end of it. Graver Tank Co. v. Linde Air Products Co., 339 U.S. 605, 607 (70 S. Ct. 854, 94 L. Ed. 1097) (1949). File wrapper estoppel is not available as a defense in such situations since the patentee here is not attempting to recapture what was abandoned before the Patent Office. Weaver Mfg. Co. v. Bear Mfg. Co., 109 F.2d 112, 115 (7th Cir.1940).

'4. The Court concludes that as a matter of law, the requisite invention was present in the Kwitek patent, and that the patent is valid. 35 U.S.C. 101-103; Graham v. (John) Deere (Co. of Kansas City) 383 U.S. 1 (86 S. Ct. 684, 15 L. Ed. 2d 545) (1960). Copying reinforces our conclusion of validity. Charles Peckett (Peckat) Mfg. Co. v. Jacobs, 178 F.2d 794, 801 (7th Cir. 1949); Bryan v. Sid W. Richardson, Inc., 254 F.2d 191, 196 (5th Cir. 1958).'

The court in its findings summarized the invention as follows:

'The patent thus describes a perforating structure which has one rotating roll, the bedroll, around which the web is partially wrapped, equipped with thin, flexible blades. These blades are rigidly supported along one edge. Further, these bedroll blades during operation contact skewed stationary blades and while being bent into associated recesses, transversely perforate the web. The stationary blades are also described in the claims as rigidly mounted or supported in a blade-supporting block or knife holder, which block or knife holder does not move or rotate during perforation. Thus, both blades being rigidly supported, one must bend to overcome the interference.'

The court found (finding 29) that defendant on September 17, 1960, in response to requests for admissions of fact, 'admitted that all elements of claim 1 were present in defendant's machine with the exceptions that the anvil blades were (1) stationary and (2) rigidly supported.'

Defendant on brief here states the contested issues thus:

'1. Did the Trial Court err in failing to interpret the claims in the light of the actual cutter shown in the patent and as required by the crowded prior art and the patent file history; and, instead, in stretching the claim language in an unwarranted manner to include Defendant's different cutter?

'2. Did the Trial Court, here, err in its holding of infringement without any determination of what the actual claimed invention was and whether it satisfied the required tests to be a patentable invention?'

Defendant in its summary of arguments states:

'The six patent claims in suit are each directed to a web cutting apparatus having relatively movable rotary and stationary intersecting blades; and these claims, after cancellation of some twenty-seven claims over prior patents, were specifically restricted by Kwitek to the blades both being rigidly supported so as to obtain their allowance (AII 1453-1455). The examiner in rejecting claims, that were later cancelled said: 'These claims do not define both the stationary and rotatable blades as being regidly supported (AII 1453).' Kwitek immediately thereafter responded by cancelling all the claims and presenting claims that became patented claims and stated: 'It is believed that the claims as now presented also include all the suggestions made in the last Office letter."

'As was found by the Patent Office and acceded to by Plaintiff in the cancellation of numerous and repeatedly rejected claims (such as claim 28, AII 1440), the basic combination of intersecting rotating yieldable and non-rotating or stationary blades was long old in the art and, hence unpatentable to Plaintiff. It was only after such rejections and in response to the Examiner's requirement to limit the claims to both or all blades being 'rigidly supported' (AII 1453), as shown by Kwitek that claims so limited were presented and allowed (AII 1461).

'The broadened interpretation of its claims, successfully ruged by Plaintiff in the trial, and in its contention that Defendant's spring supported anvil blade structure is an infringement, results in such claims being unpatentable and invalid under both Section 103 and Section 112.'

Based on defendant's trial theory, the court found (finding 30):

'The only difference between the machine disclosed in the patent and the accused structures relied upon by the defendant as avoiding infringement has been succinctly described by the defendant as follows:

'In the patented structure, the anvil balades are solidly clamped against the beam portion of the machine frame, whereas in the accused structure a transverse shaft pivotally supports the blades through an anvil casting carried by the shaft, a spring being interposed between the anvil and the frame beam.' It is the position of the defendant that since springs are interposed between the casting holding the stationary blades and the beam, those blades are not rigidly supported or mounted. Thus, the sole question here presented is whether the stationary or anvil blades in the accused machines are rigidly mounted or supported as required by the claims in issue. Defendant has not urged any further distinction based on the word 'stationary.' Defendant's own instruction sheets (PX 49) characterize the anvil blades as stationary.'

The case is unusual in that of the voluminous findings made by the trial court, only a few are criticized as not being substantially supported; in fact, only six are mentioned in defendant's brief. Defendant's criticism in the main is directed at what the court failed to find rather than at what it did. Particular criticism is made of the court's failure to make findings relating to the prior art so as to show the scope of the claims which it is asserted is a prerequisite in the determination of infringement.

The most controversial finding is as follows (finding 46):

'During the trial, the scope and content of the prior art was brought to the Court's attention even though invalidity of the patent has not been urged. During oral argument, after it had been pointed out to the Court that the last words in defendant's Second Answering Brief were 'Validity of the Kwitek patent has been disposed of as an issue here and the owners of the patent possess a legal monopoly', the Court asked counsel for defendant whether the Court had to go into validity. The answer was unequivocal, 'No, sir, you do not.' (Transcript oral argument page 66). This was followed by the Court's question, 'We are down to one issue then, infringement?'. To this, counsel for defendant replied, 'Correct'.'

Defendant attacks this finding as incomplete. It argues in effect that it made only a conditional concession of validity, dependent upon the court's decision on infringement. In other words, if the court held non-infringement, validity was conceded; if it held infringement, the concession was not applicable. While such a position might be deduced from counsel's statement made at the beginning of the trial, it appears to have been abandoned, as shown by a colloquy which took place at the closing oral argument.3

Early in the trial, defendant's counsel stated to the court, 'As you know, counsel (plaintiff's counsel) in his opening statement, and I am inclined to agree with him, says this a very simple case and it all comes down to this matter of the word 'rigid'. I think we all agree on that.' On another occasion he stated, 'Excuse me, Your Honor. If we want to move this along as Mr. Fallon said, with the exception of one limitation, it's been stipulated and admitted that the Defendant's machine responds to these various elements and the only argument on infringement is the question of these pivotable anvils versus the patent's rigid anvils. The claim is a long one, and if Mr. Fallon wants to move us along, this is all agreed to.'

In resume, defendant here conceded that the claims in suit were allowed only when they were limited to disclose stationary blades 'rigidly supported'; admitted in response to requests for admissions of fact that all elements of the claims were present in defendant's machine other than that which called for stationary blades 'rigidly supported'; stated in closing argument, 'Nobody that I have heard of, and I have pretty well investigated it, has made the socalled rigid anvil perforator.' On another occasion during the course of the trial defendant stated, 'With the exception of one limitation, it's been stipulated and admitted that the defendant's machine responds to these various elements, and the only argument on infringement is the question of these pivotable anvils versus the patent's rigid anvils.'

The court in its opinion (274 F. Supp. 374) stated, 'The sole question here presented is whether the stationary or anvil blades in the accused machines are rigidly mounted or supported as required by the claims in issue,' and incorporated this statement in its findings of fact. The feeble challenge which defendant makes to this statement is devoid of merit.

Defendant's brief is permeated with the argument that the court improperly construed the claims so as to make them read upon the accused structure. Typical is its statement, 'Under well established legal authorities, hereinafter referred to, the Court thus committed serious error when it construed the claims broadly and so as to embrace what was cancelled and dedicated in the file history of the patent.' Posing the question differently (see contested issue 1, supra), it accuses the court of erring 'in stretching the claim language in an unwarranted manner to include Defendant's different cutter.'

Of the numerous cases cited in support of this argument we need mention only Graham et al. v. John Deere Co. of Kansas City et al., 383 U.S. 1, 86 S. Ct. 684, 15 L. Ed. 2d 545, and Fife Mfg. Co. v. Stanford Engineering Co., 299 F.2d 223 (CA-7). In Graham (page 34, page 702 of 86 S.Ct.) the court stated:

'Here, the patentee obtained his patent only by accepting the limitations imposed by the Examiner. The claims were carefully drafted to reflect these limitations and Cook Chemical is not now free to assert a broader view of Scoggin's invention.'

In Fife (299 F.2d page 226) this court stated:

'It is well settled that a patent owner may not apply a narrow construction to his claim to avoid the prior art and then apply a broad construction to include an accused device.'

Defendant's argument is fallacious because it is based upon a non-existent premise. Consequently, the cases cited in its support are not in point. The court did not construe the claims 'broadly'; neither did it 'stretch' the claim language. On the contrary, it applied the claims literally and found that they read directly on the accused structure. As previously shown, defendant conceded that all elements of the claims read directly upon the accused structure other than that which called for anvil blades 'rigidly supported.' As to this element the court found, 'Under these circumstances, the Court finds that anvil blades in the accused structures are rigidly supported,' and 'the defendant's machines literally meet the claims and perform in a fashion identical to the patented machines when cutting a web as specified in the claims.'

Closely akin to defendant's argument lastly discussed is its contention that the court's findings are fatally defective because it failed to analyze the prior art so as to determine that which was covered by the invention and that which remained in the public domain. The record reveals, as is disclosed by the court's findings, that it went into much detail in considering the prior art patents offered by defendant. True, it appears that this was more related to the issue of infringement than to the scope of the invention. It seems to us that defendant is in an awkward position to complain, because it was responsible for the posture on which its case was presented and decided. Relative to this situation defendant on brief states, 'In this respect the record shows Defendant believed its defenses of non-infringement and file history estoppel were so strong that they should be allowed; and in which event the Court would not have to pass on validity.' We do not know, of course, what defendant believed, but we find it reasonable to think that it recognized the futility of defending on the ground of non-validity and chose instead to make its defense solely on non-infringement and file history estoppel.

We see no point in citing or discussing cases which announce the general rule that the court on the issue of validity must resort to the prior art and the patent history where the scope of the invention is in question. This is not such a case. If there is anything certain here, it is that the claims were allowed over the prior art, with the limitation that the anvil blades be 'rigidly supported.' This much the defendant time and again admitted. It not only conceded validity with the claims so limited, but advised the court that 'Nobody that I have heard of, and I have pretty well investigated it, has made the so-called rigid anvil perforator.'

In view of the situation presented, we are not convinced that the court committed reversible error in not further exploring the prior art and failing to make a specific finding on the defense of non-invention. In its conclusions the court did state that as a matter of law the Kwitek patent was valid. Of course, we recognize that it has been held that a court cannot be estopped from inquiring into the issue of validity by agreement or conduct of the parties, but we think such cases are distinguishable on their facts from those here. Defendant relies particularly on Haughey v. Lee et al., 151 U.S. 282, 14 S. Ct. 331, 38 L. Ed. 162 and Paramount Publix Corp. v. American Tri-Ergon Corp., 294 U.S. 464, 55 S. Ct. 449, 79 L. Ed. 997. In both of these cases it was held that the inconsistency of a defendant's prior attempt to procure a patent on the same subject matter afforded no basis for an estoppel to attack the validity of the patent which he was charged to infringe. In Paramount, the court stated (page 477, page 445 of 55 S.Ct.):

'However inconsistent this early attempt to procure a patent may be with petitioner's present contention of its invalidity for want of invention, this Court has long recognized that such inconsistency affords no basis for an estoppel, nor precludes the court from relieving the alleged infringer and the public from the asserted monopoly when there is no invention. Haughey v. Lee, 151 U.S. 282, 285 (14 S. Ct. 331, 38 L. Ed. 162).'

We have already expressed our agreement with the trial court's conclusion that the sole question presented on the issue of infringement 'is whether the stationary or anvil blades in the accused machines are rigidly mounted or supported as required by the claims in issue.' Defendant sought to escape infringement on the basis that in its device springs are interposed between the stationary blades and the beam and, consequently, the blades are not rigidly supported. Of the court's voluminous findings we set forth such portions as are particularly pertinent to the crucial issue of infringement. The court found:

'36. It was the position of the plaintiff that the anvil springs due to 'preload' or precompression, do not perform as springs during 'normal operation', but perform as solid blocks so that there could be no movement either about the shaft or into the clearance. The plaintiff further urged that the defendant's interposition of springs, with their ability to compress when abnormal conditions arise, constituted at best an improvement over the patented structure and does not avoid infringement. The plaintiff does not question the fact that the springs can be further compressed during what it terms 'abnormal' conditions, such as the presence of wads or wrap-ups, so as to permit pivoting of the anvils about the shaft as urged by defendant. The operation characterized by plaintiff as 'normal' has been described by the defendant as 'ideal'. When we use the term 'normal' operation, we refer to an operation in which the web is moving through the perforating unit and being perforated. In so doing, we do not imply that there is not ever present the possibility of wads and wrap-ups at the perforator which possibility, if occurring, necessitates stoppage of the unit whether it be of plaintiff's or defendant's construction, this being shown by movies (DX 286 and 287) and on which various witnesses testified. In the movies shown to the Court, wads going through the perforator required the machine to be shut down and cleaned, as testified to by defendant's own engineer, Besserdich. (R. 882).

'37. The parties are in agreement that the springs associated with each anvil exert eighty pounds of force. This force is derived from the compression of the springs effected by the position of the adjusting screw. The springs are compressed from their free length and held in this condition by the cooperation of the adjusting screw and stop button. This arrangement is denominated by engineers as spring 'pre-load' and means, in effect, that the pre-load must be exceeded (here eighty pounds) before further spring compression can be effected. If there is no further compression during normal operation, the springs function as a solid bar and the anvil blades are rigidly mounted. It is clear that any movement or pivoting of the anvils (and thus the anvil blades) in defendant's accused machines during normal operation must be caused by interference between the thin, flexible bedroll blades and the anvil blades as the bedroll rotates. It is also clear that the force from the roll blades necessary to produce such movement must be sufficient to overcome the force exerted by the springs which hold the anvils in position against the adjusting screw contacting the stop button. * * * All the force the roll or rotating blade could exert during contact at normal interferences was 22 1bs. The Court participated in a demonstration using a fish scale and noted that, even with a mechanical advantage, it was necessary to exert a pull of 65 1bs. on the No. 1 anvil in order to cause movement. The Court finds that the bedroll blades exert insufficient force to overcome the force exerted by the springs and thus are incapable of causing movement of the anvils during normal operation, and this irrespective of what pivot point the movements were based on.

'43. Barlament and Bradley, engineers who had experience in operating both plaintiff's and defendant's machines testified that in normal operation, it was possible to perforate a sheet without anvil movement. This leaves, therefore, only the question of whether the anvil blades are resiliently supported or mounted by reason of the fact that movement of the anvils is concededly permitted by the springs during abnormal operation. This, in essence, is the approach and opinion of Dr. Jacobsen. Dr. Jacobsen based his opinion that the anvils were resiliently mounted on the fact that they have the ability to move, or as he phrased it, 'the ability to roll with the punches' (R. 1178). In the end he admitted that his tests proved nothing with respect to motion of the anvils during normal operation but maintained that the mere presence of springs made the anvils resiliently mounted.

'44. In support of this latter approach, the Court finds that there was no evidence to show that the springs were interposed in defendant's machine to permit movement of the anvils and anvil blades during normal perforating operation or to perform any function during such operation. Such an advantage or purpose was not shown to have been advanced by the defendant as a desirable feature when the perforators were introduced. Rather, the springs are described in the defendant's early literature as holding the anvil in position and no statement is made that the anvil movement during perforation occurs or is desirable.

'45. While a spring is unquestionably resilient, it can function as a rigid bar and the springs in defendant's machines do so function until their pre-load is overcome, a condition which does not occur while the structures are performing the function for which they are designed, that is, perforating paper. When the springs cease to operate as a rigid bar, that is, when their latent resilient capabilities are called into play by overload conditions, the structures cease to perform their intended function since the machines must be stopped to correct the trouble in the web being fed through them. Under these circumstances, the Court finds that anvil blades in the accused structures are rigidly supported or mounted within the meaning of the claims in issue. The Court further finds that a spring mounting or support can, by its very nature, be considered rigid in situations where the spring functions as a rigid member holding the anvil blade in cutting position and preventing movement while perforation is taking place. Further, the Court finds that plaintiff is not improperly attempting to interpret its patent as covering a mode of operation (rather) than a structure. The defendant's anvil support is rigid for all practical purposes during operation of the machine and the interposition therein of a member having resilient capacities not utilized during perforation does not avoid infringement. Such a member may be an improvement over the corresponding member in the patent device as by providing automatic pivoting of the anvils on overload, overload releases being common safeguards. The feature of employing movable anvils is also taught in the Kwitek patent and covered by claim 9. It is expected that changes or improvements will be made by later comers. But this does not alter the fact that the defendant's machines literally meet the claims and perform in a fashion identical to the patented machines when cutting a web as specified in the claims.'

Defendant persists in its argument that the court's ultimate finding of infringement was based upon an erroneous interpretation of the 'rigidly supported' stationary cutting blades of the claims in suti so as to read upon the spring mounted pivotal anvil of defendant's structure. We again reject this argument. The court evaluated defnedant's structure alleged to infringe and found that in normal operation its blades were rigidly supported, as called for in the claims. Defendant contends that the court improperly characterized its operation as 'normal' and 'abnormal'. We think the court in the findings above quoted satisfactorily answered this contention.

In considering the court's findings relative to the issue of infringement, little if any aid might ordinarily be expected from case law because of the difference in their factual situations. However, two cases called to our attention merit consideration because of their factual similarities, as well as the reasoning employed by the courts. Weaver Mfg. Co. et al. v. Bear Mfg. Co., 109 F.2d 112 (CA-7), and Farrington v. Haywood, 35 F.2d 628 (CA-6).

In Weaver, the patentee's claim was limited to a disclosure of a rigid connection of the piston rod to a platform. Defendant contended that in its alleged infringing structure there was no rigid connection between the piston rod and the platform. This court, in denying defendant's contention that it escaped infringement because of the 'rigid connection' element, stated (page 115):

'The word 'rigid' has a variety of meanings, which in our opinion would not support appellant's contention. However, we assume that the word 'rigid' means stiff and inflexible, but we think that a rigid connection does not necessarily mean a fastening together. There may be a rigid connection by a mere contact. Appellant is quite right in stating that in its device there is no connection at all when the tester is not in use. However, when the contact is made between these headers by the thrust caused by the locked wheels of the automobile, for all practical purposes the connection is just as rigid as if the two headers were welded together. It is quite true that appellant has a different method of restoring the platform to its original position after the test has been made, but with that we are not concerned and it is not necessary to describe it here.'

In Farrington, the patent was on a stirrer to be inserted into a paint berrel. By action of the Patent Office, the claims were limited to a one-piece stirrer. The defendant made a two-piece stirrer, hinged together so as to operate on a different principle under the abnormal condition of accumulating a large amount of sediment. In holding the patent infringed, the court stated (35 F.2d page 631):

'There is a substantial identity of form and method of operation with only 'such variations as are consistent with its being in substance the same thing.' * * * It will therefore infringe at some stage of its operation. It is unnecessary that it infringe at all stages.'

Farrington was cited with approval by this court in Highway Appliances Co. v. American Concrete Expansion Joint Co. et al., 7 Cir., 93 F.2d 113, 118.

As noted in the beginning, this was a protracted case, presided over by Judge Tehan from its inception to its conclusion. The trial consumed three weeks, at the conclusion of which oral argument was heard for an entire day. Both the patent structure and that alleged to infringe were operated in the presence of the court. Numerous expert witnesses demonstrated and testified as to their operation. The court witnessed movies of the structures in operation. The record is convincing that Judge Tehan had an acute understanding of the structures and their manner of operation.

In Graver Tank & Mfg. Co., Inc. et al. v. Linde Air Products Co., 336 U.S. 271, 69 S. Ct. 535, 93 L. Ed. 672, in discussing Rule 52(a) the court stated (page 274, page 537 of 69 S.Ct.):

'To no type of case is this last clause more appropriately applicable than to the one before us, where the evidence is largely the testimony of experts as to which a trial court may be enlightened by scientific demonstrations.'

'The rule requires that an appellate court make allowance for the advantages possessed by the trial court in appraising the significance of conflicting testimony and reverse only 'clearly erroneous' findings.'

In our view, the findings of the court as they relate to the issue of infringement are adequately supported by the record. Certainly there is no reasonable basis upon which we could hold that they are 'clearly erroneous.' Such being the case, the findings support the judgment of infringement.

We deem it unnecessary to set forth the findings which describe in detail the ten-year effort made without success by plaintiff, defendant and their competitors to develop an apparatus which would produce toilet and other paper webs satisfactory to their customers. A summary of such activities is shown in the District Court's opinion. Paper Converting Machine Co., Inc. v. FMC Corp., 274 F. Supp. 372. It is sufficient here to state that the problem which vexed the industry for many years was solved by the apparatus disclosed in the patent in suit. Customers of plaintiff's competitors, including defendant, demanded a quality of product equal to that produced by plaintiff. We think the record clearly reveals that defendant met this response by developing the apparatus alleged to infringe.

Defendant contends that it was not an infringer because its apparatus was a patentable improvement, as shown by Patent No. 2,986,058, issued May 30, 1961, to O. Besserdich (its engineer). No case is cited which supports this argument. This court in Weaver Mfg. Co. et al. v. Bear Mfg. Co., 109 F.2d 112, considered the same contention, and in response thereto stated (page 114):

'There seems to be no doubt that a patent may issue for improvements on an existing patented device. Such subsequent patents, however, would not in any way relieve such patentee from liability for infringement of the prior patent if used as an element in the later patent.'

The court in its findings (above quoted) suggested that defendant's apparatus might be an improvement, and plaintiff on brief here concedes as much; however, it is a rule well established that a mere improvement does not avoid infringement. Temco Electric Motor Co. v. Apco Mfg. Co., 275 U.S. 319, 48 S. Ct. 170, 72 L. Ed. 298; Nordberg Mfg. Co. v. Woolery Machine Co., 7 Cir., 79 F.2d 685, 692; Highway Appliances Co. v. American Concrete Expansion Joint Co. et al., 7 Cir., 93 F.2d 113, 118. In Temco, the court stated (275 U.S. page 328, 48 S.Ct. page 173):

'It is well established that an improver cannot appropriate the basic patent of another, and that the improver without a license is an infringer, and may be sued as such. (Citing cases.)'

Lastly, we agree with the trial court's conclusion that file wrapper estoppel is not available as a defense. As previously shown, we have approved the trial court's determination that the claims in suit read literally upon the accused structure. Thus, there is no occasion to resort to the doctrine of equivalents. Of the numerous cases called to our attention, the one most relevant is Smiths America Corp. v. Bendix Aviation Corp., 140 F. Supp. 46 (D.C.D.C.), affirmed 101 U.S.App.D.C. 299, 248 F.2d 621, 626. In that case the District Court cited many cases in support of the statement (140 F.Supp. page 54), with which we agree:

'But the Court is of the opinion that no question of File Wrapper Estoppel arises unless it can be shown that it is necessary to go beyond the ordinary and accepted meaning of the words in the allowed claims, and not even then unless it can be shown that plaintiffs must resort to the doctrine of equivalents. Therefore, since the Court has determined, as stated above, that the claims in plaintiffs' patents read directly on defendant's sextant, it is unnecessary for plaintiffs to resort to the doctrine of equivalents. Thus the authorities cited by defendant under its defense of File Wrapper Estoppel have no applicability in this case.' The judgment appealed from is

Affirmed.