OPINION OF THE COURT
ADAMS, Circuit Judge.
This appeal requires us to decide the proper role for the patent law concept of “synergism,” the proposition that to serve as the subject of an enforceable patent a combination of known elements must somehow exceed the sum of its parts. Additionally, we are faced with questions regarding [537]*537the priority to be accorded a foreign patent application and the scope of disclosure in a domestic application. Because we believe that the district court — which decided that the patent in question was not enforceable — erred in imposing a synergism requirement, we vacate and remand.
I.
On August 27, 1974, the United States Patent Office issued Patent No. 3,831,502 (“the ’502 patent”) to plaintiff Rengo Co., Ltd., as assignee of the named inventor, Masatero Tokuno. The patented apparatus includes portions of a “corrugator,” a machine that manufactures the three-ply paperboard from which the familiar cardboard box is made.
Production of corrugated paperboard begins with the introduction of three webs of paper into the machine. The machine initially corrugates (i. e., flutes) one web, and the flutes of the opposite surfaces of this sheet are glued to the two flat webs of single-ply paper. The resulting three-ply web then passes over a heating bed where the glue is set and dried. The paperboard next travels to the “slitter-scorer” complex. This series of instruments first slits the web along the running direction into two or more bands of selected width, and then, in the operation known as “scoring,” imposes longitudinal creases on each of these strips in order to facilitate the subsequent construction of the boxes. Finally, the web passes to a rotary shear that cuts across the paperboard, producing box blanks of predetermined length.
By changing the pattern of slitting and scoring, a single corrugator can manufacture cardboard boxes of various sizes and configurations. Moreover, a corrugator may be equipped with more than one slitter-scorer unit. As a result, while one unit is in operation the other is accessible for adjustment to accommodate the next run. Before the development of the subject invention, an operator of a corrugator who desired to effect an “order change,” /. e., a change in the slitting and scoring pattern, would use the rotary shear to make a transverse cut in the paperboard web. The portion of the web downstream of the transverse cut would pass through the slitting and scoring rolls then in operation. When the old order was completed, the portion of the web behind the point of severance would proceed to the slitting and scoring rolls arranged in the desired new pattern. An order change effected by this method requires the operator to stop the corrugator completely, or at least to slow the running of the paperboard web through the machine. A decrease in the productivity of the machine thus inevitably attends an order change.
The subject matter of the patent in suit was designed to improve the efficiency of conventional corrugators by performing the slitting and scoring operation without suspending or reducing the running speed of the web during an order change. The patented apparatus comprises a conventional rotary shear, located upstream of two conventional slitter-scorer units. The combination also contains a movable feeding plate that can direct the web to an upper path, which leads to one of the slitter-scorer units, to a lower path, which leads to the other slitter-scorer unit, or to an intermediate path, which flows between the two slitter-scorer units. To effect an order change, the rotary shear makes a single transverse cut in the web, and the feeder plate instantaneously diverts the leading edge of the severed web to the slitter-scorer not previously in use. The difference in conception between conventional corrugators and corrugators equipped with the patented device can be put simply: In the conventional device, the paperboard web travels along a single pathway; during order change, a new slitter-scorer unit moves into the path of the board to engage the web. The patented device, by contrast, changes the path of the web to meet the new slitting and scoring unit.
Rengo initially developed a corrugator capable of instant order change in Japan. It installed machines employing the new patented apparatus at two of its Japanese plants in January 1972, and filed Japanese [538]*538patent applications in February and April of the same year. Rengo filed its application for an American patent in January 1973.
During the middle 1970’s, the Langston Division of defendant Molins Machine Company, in response to market pressures and consumer demands, set about to develop a corrugator capable of performing instant order change. In 1976, two representatives of Molins visited Japan and while there observed the operation of a Rengo corrugator. In December 1977, Langston began development of corrugator apparatus similar to the device which Rengo had already patented. Like the Rengo mechanism, the Langston version achieved instant order change by diverting the flow of paperboard from one set of slitters and scorers to another. The principal difference between the two devices is that Langston employs only two flow paths, or “board lines,” whereas Rengo utilizes three. Langston offered its instant order change system for sale in early 1978, and installed a prototype later that year at the Louisville plant of the Mead Corporation. Another corrugator employing the Molins double board line system went into commercial operation in December 1979.
A.
In August 1978, Rengo filed a complaint in the District Court for the District of New Jersey, alleging that Molins had infringed claims 1 and 24 of the ’502 patent. In its answer, Molins denied infringement and requested that the court declare the ’502 patent invalid and, in any event, not infringed. In addition, Molins asserted by way of a counterclaim that Rengo had violated the antitrust laws.
After a four day trial, the district court, finding the ’502 patent invalid, entered judgment in favor of Molins. The court recognized that an inquiry into validity requires comparison of the patented device with the state of the art before the invention. In this case, the district court, in order to make the comparison, had first to determine whether the Japanese filing date, as opposed to the American filing date, should be considered the date of invention. After resolving this issue in favor of Rengo, the court nonetheless went on to hold the ’502 patent invalid. The court’s disposition of this issue rested to a considerable extent on its finding that the patented device did not demonstrate a “synergistic result.”
Although an invalid patent cannot be infringed, the court next considered Rengo’s infringement claims, reasoning that a resolution of these questions might expedite final disposition of the case if the court of appeals reversed the district court’s ruling on validity. Finding that the Molins and Rengo machines were equivalent in means, operation, and result, the court held that the ’502 patent, if valid, was infringed. Finally, the court considered and dismissed Molins’ antitrust claim.
Rengo appeals from the court’s finding that the ’502 patent is invalid, while Molins cross-appeals from the court’s determination that the patent, if valid, was infringed.
II.
Initially, we confront a potential barrier to our jurisdiction. Rengo predicated its complaint on a theory of infringement, as that term is defined in 35 U.S.C. § 271(a): “Whoever without authority makes, uses or sells any patented invention, within the United States during the term therefor, infringes the patent.”1 At the time Rengo filed its complaint, however, Molins had done no more than advertise and solicit orders for its new instant order change apparatus; it had not yet begun the manufacture or sale of the equipment. Molins contends that its conduct at the time Rengo instituted the lawsuit did not rise to the level of infringement within § 271(a). A jurisdictional prerequisite to the maintenance of the infringement action was there[539]*539fore lacking, and, Molins asserts, both the district court and this Court lack authority to adjudicate Rengo’s claim.
Without deciding whether the district court had jurisdiction over Rengo’s original cause of action,2 we conclude that the court had power to consider the issues of patent validity and infringement. Regardless of deficiencies in the complaint, Molins’ assertion of a counterclaim, in which it requested the court to declare that the ’502 patent was invalid and that it was not infringed, conferred jurisdiction on the district court over both questions.
The actual manufacture, use, or sale of an allegedly infringing device is not a condition precedent to a suit seeking declaratory relief against the holder of the patent. In order to maintain an action, it is sufficient that the potential infringer have the “immediate intention and ability to practice the invention.” 3 In Wembley, Inc. v. Superba Cravats, Inc., 315 F.2d 87 (2d Cir. 1963), the plaintiff requested a declaration that the defendant’s patent was invalid or, alternatively, that plaintiff’s proposed use would not infringe the patent. Although the plaintiff had not yet begun production, the Second Circuit held that this, without more, would not bar the suit. The court reasoned that “it would be economically wasteful to require a plaintiff to embark on an actual program of manufacture, use or sale which may turn out to be illegal.” Commending this analysis in Super Products Corp. v. D P Way Corp., 546 F.2d 748, 753-55 (7th Cir. 1976), the Seventh Circuit held that a party possessing the “apparent ability and a definite intention ... to manufacture and sell a product similar to the one described in [the contested] patent” can press for declaratory relief, provided it has a reasonable apprehension that it will face an infringement suit if it continues the activity in question.
In the present case, no jurisdictional impedimént barred adjudication of Molins’ request for declaratory relief concerning patent validity and infringement. By advertising and soliciting orders, Molins manifested a “definite intention” to manufacture the accused device. Moreover, Molins’ installation of an instant order change system in November 1978, just two months after it filed its original answer and counterclaim, establishes its “apparent ability” to produce a possibly infringing device. In addition, Rengo’s initiation of an action for infringement left no doubt that Molins would be forced to defend itself if it continued to manufacture and sell the accused machinery.
Notwithstanding the question of jurisdiction over the original complaint, then, the district court had jurisdiction over the Molins counterclaim. And because the counterclaim raised the same issues that the complaint presented — whether the ’502 patent was valid and, if valid, whether it would be infringed by the Molins device — the court had power to decide those issues. Further, a jurisdictional defect in the complaint will not preclude adjudication of a counterclaim over which the court has an independent basis of jurisdiction. As we stated in National Research Bureau, Inc. v. Bartholomew, 482 F.2d 386, 388-89 (3d Cir. 1973) (per curiam): “Once an issue with independent subject matter jurisdiction is before the court and jurisdiction over the parties has been perfected, it must be allowed to proceed to a conclusion pursuant to the rules of civil procedure like any other routine federal claim . .. [w]here, as here, jurisdiction is independent, the counterclaim must be allowed to proceed without [540]*540regard to the fate of the original claim.” 4 Thus we need not decide whether Rengo’s complaint is cognizable in federal court. It is enough to say that the district court, having an independent basis for jurisdiction over Molins’ counterclaim, had the authority to adjudicate the issues of patent validity and infringement raised therein.
III.
The most significant question in this case is whether the district court tested the ’502 patent against the appropriate standard. Rengo argues that the court, in requiring a demonstration of “synergism,” improperly supplemented the statutory criteria for patent validity. To resolve this issue, we first must ascertain whether the synergism test ís mandated either by statute or by recent Supreme Court decisions. After examining the relevant precedents, we will consider whether such a test, even if not required, might nonetheless comport with modern principles of patent jurisprudence.
Between 1790 and 1952, the American patent statutes expressly referred to only two prerequisites to the grant of a patent: novelty and utility.5 These elements, presently codified at 35 U.S.C. §§ 101 and 102,6 did not exhaust the standards of patent validity, however. As early as 1851, the Supreme Court added a third condition, denominated “invention,” for the issuance of a patent. The Court noted in Hotchkiss v. [541]*541Greenwood, 52 U.S. (11 How.) 248, 267, 13 L.Ed. 683 (1851): “Unless more ingenuity and skill . . . were required . . . than were possessed by an ordinary mechanic acquainted with the business, there was an absence of that degree of skill and ingenuity which constitute essential elements of every invention. In other words, the improvement is the work of the skillful mechanic, not that of the inventor.”
For the next one hundred years, elaboration of the invention requirement devolved on the judiciary, as the patent statute continued to embody the dual standard of novelty and utility.7 During this period, the courts developed so-called “negative rules of invention,” — decisions that certain types of improvements or changes did not exhibit the requisite degree of inventiveness.8 Significantly for our purposes, the Court stated on several occasions that a combination of known elements would not be patentable “unless some new and useful result, an increase of efficiency, or a decided saving in the operation, is clearly attained.”9 Although the Court originally focused on a “new and useful result, this test developed into a standard applicable to the component parts, as well as to the result achieved by the new device. Thus, in A&P Tea Co. v. Supermarket Corp., 340 U.S. 147, 151-52, 71 S.Ct. 127, 129-30, 95 L.Ed. 162 (1951), the Court stated: “The conjunction or concert of known elements must contribute something; only when the whole in some way exceeds the sum of its parts is the accumulation of old devices patentable.... A patent for a combination which only unites old elements with no change in their respective functions . . . obviously withdraws what already is known into the field of its monopoly and diminishes the resources available to skillful men.” Finding that the combination before it was “wanting in any unusual or surprising consequences,” and that the old elements which made up the device did not “perform any additional or different function in the combination than they performed out of it,” the Supreme Court in A&P Tea held the device non-patentable.
Partially in response to the judicially created “negative rules of invention,” 10 Con[542]*542gress amended the patent laws in 1952, retaining the requirements of novelty and utility, but adding for the first time a requirement of invention. Congress decided, moreover, to “start fresh semantically and to promote uniformity”11 by imparting to the invention requirement a new rubric— “non-obviousness.” Specifically, Section 103, 35 U.S.C. § 103, states that a patent will not issue “if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.”
In Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966), the Supreme Court established a procedure for determining whether Section 103 is satisfied. While the Court acknowledged that patent validity is a question of law, it also observed that the Section 103 issue “lends itself to several basic factual inquiries:”
Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art maintained. Against this background, the obviousness or non-obviousness of the subject matter is determined.
Id. at 17, 86 S.Ct. at 694. In addition to the three factors to be accorded primary significance, such “secondary considerations” id., as commercial success, long felt but unsolved needs, and failure of other attempts to develop a comparable device, might be relevant as indicia of obviousness, id. at 17-18, 86 S.Ct. at 693-94.
Significantly, the Supreme Court in Graham seemed to regard Section 103 as abolishing the judicially fashioned “negative rules of inference,”12 and replacing them with a unitary inquiry into obviousness. In the Graham case itself, the Court examined the conflicting approaches of two circuits, which had applied different “negative rules" to test the patentability of a combination of old elements. The Eighth Circuit, reflecting Hicks v. Kelsey, 85 U.S. (18 Wall.) 670, 673, 21 L.Ed. 852 (1874), had required a “new or different result,”13 while, in an earlier decision, the Fifth Circuit had required merely the production of “an old result in a cheaper and more advantageous way.” 14 The Supreme Court, disapproving both standards, directed that courts should inquire solely into whether the improvement in question would have been obvious to one skilled in the art.
In the ease at hand, the district court adhered to the Graham decision by identifying the scope of the prior art, the differences between the art and the '502 patent, and the relevant level of skill. The court, however, appeared to supplement the Graham analysis with the “negative rule” that applied to. combination patents before the adoption of Section 103. The court introduced its discussion of Section 103 by observing that “additional aspects of nonobviousness” come into play when a combination patent is at issue. It identified these “additional aspects” as a requirement that the device produce a “synergistic result” in which “the whole in some way exceeds the sum of its parts.” The court concluded that the ’502 patent could not survive scrutiny under Section 103, inasmuch as “each element of the combination performs in the same manner and with the same result as in the prior art machines.”
Rengo maintains that the district court, by imposing a requirement of “synergistic result,” revitalized one of the “negative rules of invention” discredited in Graham. In so doing, Rengo argues, the court subjected the ’502, patent to. an unduly rigorous standard, and the resulting finding of invalidity cannot stand. We agree with Rengo that Graham establishes the standards gen[543]*543erally applicable to patents alleged to violate Section 103, and that the district court engrafted onto the Graham analysis a “synergism” requirement strongly reminiscent of the pre-1952 “negative rules of invention.” While neither Section 103 itself nor the Supreme Court’s Graham decision appear to contemplate this embellishment, post-Graham developments must be understood in order to evaluate its propriety.
B.
Notwithstanding Graham, the Supreme Court, in two more recent decisions, has resorted to language which evokes the rule of Hicks v. Kelsey and A&P Tea Co. v. Supermarket Corp. In Anderson’s-Black Rock Inc. v. Pavement Salvage Co., 396 U.S. 57, 90 S.Ct. 305, 24 L.Ed.2d 258 (1969), the court struck down a claim that combined a radiant heat burner, a bituminous paving machine, and an asphalt shaper apparatus mounted on the same chassis, each of which was taught by the prior art. The Court observed that Graham established the guidelines for assessing obviousness, and, applying this analysis, concluded that “the combination was obvious to one with ordinary skill in the art.” Id. at 60, 90 S.Ct. at 307. But the Court also commented:
A combination of elements may result in an effect greater than the sum of several effects taken separately. No such synergistic result is argued here. It is, however, fervently argued that the combination filled a long felt want and has enjoyed commercial success. But these matters “without invention will not make patentability.”
Id. at 61, 90 S.Ct. at 308 (quoting A&P Tea Co. v. Supermarket Corp., 340 U.S. at 153, 71 S.Ct. at 130).
Despite the allusion to “synergistic result,” we do not understand Black Rock to supersede or to supplement Graham. To the contrary, the Supreme Court’s reference to and application of the Graham test constitutes a reaffirmation of that decision. The Court’s resort to synergism language is, we believe, best interpreted “merely to note the advent of a phenomenon which may emanate from a combination claim, without any indication by the Court that the phenomenon must be present in every case to satisfy the requirements of Section 103.” 15
In Sakraida v. Ag Pro, Inc., 425 U.S. 273, 96 S.Ct. 1532, 47 L.Ed.2d 784 (1976), the Supreme Court again considered the validity of a combination patent. As in Black Rock, the Court reiterated its fidelity to the Graham test, id. at 280, 96 S.Ct. at 1536, and, after analyzing the scope of the prior art, concluded that the particular use before it16 would have been obvious to anyone skilled in the art, id. at 282, 96 S.Ct. at 1537. Again as in Black Rock, the Court invoked the rhetoric of synergism, stating that the combination in issue could not “be characterized as synergistic, that is, ‘resultpng] in an effect greater than the sum of the several effects taken separately.’ ” Id. (quoting Anderson’s-Black Rock v. Pavement Salvage Co., 396 U.S. at 61, 90 S.Ct. at 308). The background of the Sakraida decision suggests, however, that the Court did not intend to reinstate the “negative rule of invention” formerly applicable to combination patents. The Fifth Circuit in Sakraida had held that the invention “[did] achieve a synergistic result,” 474 F.2d 167, 173 (5th Cir. 1973), and the Supreme Court directed its discussion of synergism to this lower court finding, see 425 U.S. at 281-82, 96 S.Ct. at 1537. In light of the Supreme Court’s endorsement of Graham, we believe its discussion of synergism expresses a re[544]*544jection of the Fifth Circuit’s holding, rather than the institution of an additional precondition to patentability.
In contrast to the district court, we doubt that the Supreme Court’s recent decisions impose distinctive barriers to the validity of combination patents.17 The Supreme Court precedents, we believe, do not resurrect the earlier “negative rules,” but suggest instead that Graham definitely states the minimum criteria for satisfaction of Section 103. Moreover, we note that our interpretation of Sakraida and Black Rock is not unique. Although there is a split among the authorities, at least four other federal courts of appeals have, in the wake of Sakraida, arrived at a similar analysis of the Supreme Court’s precedents.18
C.
While Black Rock and Sakraida do not compel adoption of a synergism requirement, neither do these decisions, by their terms, preclude such a result. Independent of these precedents, sound arguments counsel against incorporating into the patent law a special test for combination patents. Section 103 itself does not discriminate among various types of patents; its requirements, instead, apply to all applications, regardless of their subject matter. This uniformity is not accidental, for Congress, in enacting Section 103, intended to replace the mosaic of “negative rules,” many of which could be applied to only a limited range of patents, with a single standard of non-obviousness.19 It would confound the statutory design to impose on one class of patents a harsher test of patentability than the rest. This concern is particularly relevant to a standard like “synergism,” which is, by nature, difficult to formulate and apply. It is worth noting that those courts that construe Black Rock and Sakraida to embrace a synergism precondition for combination patents have not agreed on how the standard should be formulated.20
[545]*545Besides creating inconsistency in the patent laws, the synergism test suffers from several analytical defects. The Seventh Circuit has detailed these problems in Republic Industries, Inc. v. Schlage Lock Co., 592 F.2d 963 (1979), the leading decision rejecting the synergism standard. First, because synergism is linked exclusively with combination patents, the scope of the synergism standard derives from the definition of “combination.” As Judge Learned Hand once pointed out, however, no standards can be articulated for delineating a class of “combinations,” for every invention is a combination of old elements.
[T]he defendant argues that the supposed invention is no more than a substitution of materials familiar to the art in the same uses; an aggregation of which each part performs what it did before. We may concede as much arguendo, for the same may be said of every invention. All machines are made up of the same elements; rods, pawls, pitmans, journals, toggles, gears, cams, and the like, all acting their parts as they always do and always must. All compositions are made of the same substances, retaining their fixed chemical properties. But the elements are capable of an infinity of permutations, and the selection of that group which proves serviceable to a given need may require a high degree of originality. It is that act of selection which is the “invention” and it must be beyond the capacity of commonplace imagination.
B. G. Corp. v. Walter Kidde & Co., 79 F.2d 20, 21-22 (2d Cir. 1935).21
Assuming the definitional hurdles can be overcome, synergism, whether conceptualized as a characteristic of individual parts or of the product they cooperate to produce, can rarely, if ever, exist. In virtually every mechanical patent, the constituent parts will perform their known and expected functions, and the possibility that the elements will function differently in combination than they did separately is correspondingly remote.22 Moreover, “mechanical elements can do no more than contribute to the combination the mechanical functions of which they are inherently capable;”23 thus the performance of a combination will equal the sum of the functions of its components, and we will rarely, if ever, find that “the whole in some way exceeds the sum of its parts.” 24 If applied consistently, a synergism requirement might well foreclose the validity of many, if not most, mechanical patents.
Additionally, we agree with the Seventh Circuit’s observation in Schlage Lock that the synergism standard contravenes important objectives of patent protection. By looking exclusively to the functioning of the individual components after combination, the approach is premised on “the assumption that it is always obvious to take known elements and combine them.” 592 F.2d at 971. But the selection of elements may itself be non-obvious and therefore inventive. Focusing on the performance of elements after combination, to the exclusion of the obviousness of making the combination, thus seems inconsistent with Section 103, which establishes as the standard of patent-[546]*546ability the non-obviousness of the combination “at the time the invention was made.”
D.
Because the synergism requirement is neither mandated by the Supreme Court nor commended by independent considerations, we decline to adopt it as a prerequisite to the validity of a combination patent. Instead, we regard the three-part test of Graham as applicable to combination, as well as to other patents: to ascertain the non-obviousness of a patent, a court should look to the prior art, to the differences between the claimed invention and the prior art, and to the level of skill in the industry, with appropriate weight accorded to “secondary considerations.” To the extent that the district court deviated from the Graham analysis by measuring the ’502 patent .against the standard of synergism, it committed error. Moreover, our review of the Court’s opinion does not persuade us that the error was harmless.
As noted above, the court recited various of the maxims traditionally associated with synergism: the whole must in some way exceed the sum of its parts, and at least some elements of the combination must perform in a novel or unpredictable manner. While a mere reference to synergistic aphorisms might not prejudice the holder of a patent, we are convinced that the district court, at least to some extent, substituted synergism for the Graham analysis. Specifically, it deviated from Graham by failing to consider adequately whether the specific combination of elements before it — each, admittedly, performing a known function— was obvious at the time of invention. For example, the court acknowledged that the prior art did not teach the placement of the diverter plate upstream of the cutting and grooving tools. In inferring that the development of a diverter plate located between the rotary shear and slitter-scorer units was not an inventive contribution, the judge merely recited the prior art and then stated that each element of the new combination performs its old function. The court did not explain what made this specific combination of known elements obvious to one skilled in the art. Moreover, its injection of synergism resulted in an additional departure from the Graham analysis. In declining to look to “secondary considerations” such as the failure of Rengo’s rivals to develop an adequate corrugator, the court properly noted that these factors cannot revive a patent already adjudicated unsatisfactory by Section 103 standards. But in this case, the determination of invalidity that foreclosed secondary considerations rested on the synergism standard. In short, we find that concepts of synergism so permeated the trial court’s ruling on patent validity that we cannot sustain that holding.
Occasionally, a reviewing court, when presented with a finding of patent invalidity premised on incorrect legal standards, will proceed to decide the validity issue on its own, without remand to the trial judge.25 Under the circumstances of this case, however, we believe the better practice is to allow the district court to reexamine the ’502 patent under the Graham test. The Graham guidelines contemplate a refined comparison of the prior art and the subject of the ’502 patent; the district court, having conducted a four day trial, is more familiar with, what is in this case, a complex body of art, and hence better situated than we to construct the detailed analysis which its initial reliance on synergism rendered unnecessary. For example, the district judge, in assessing the placement of [547]*547the rotary shear and diverter, may wish to consult the expert testimony and other exhibits which the parties submitted to it, but on which it did not previously rely. In addition, the secondary considerations that the court summarily set aside may merit greater weight when it examines the patent without seeking a synergistic result. We believe that the district court is the best forum for development of the factual basis on which the legal judgment concerning validity ultimately rests.
IV.
Two issues touching on validity, both of which the district court resolved in favor of Rengo, remain. The first concerns the proper date for ascertaining the state of prior art; the second, which is closely allied, concerns the formal adequacy of Rengo’s application for a patent. Although we have already determined that the district court’s judgment must be vacated, the resolution of these questions will affect the scope of the inquiry on remand. Accordingly, we find it appropriate to address them at this time.
Although the statute indicates that the obviousness of an invention should be evaluated in light of the art which existed “at the time the invention was made,” in practice the reference point generally has been understood as the date the inventor filed an application with the Patent Office. Eltra Corp. v. Basic Inc., 599 F.2d 745, 752 n.10 (6th Cir. 1979); U. S. Expansion Bolt Co. v. Jordan Indus., Inc., 488 F.2d 566, 568 n.3 (3d Cir. 1973). Section 119 establishes an exception to this general rule, however. If an inventor applies for a patent in a foreign country, and files an application in the United States within one year, the American application “shall have the same effect as the same application would have if filed in this country on the date on which the application for patent for the same invention was first filed in such foreign country.” Consequently, an American application which falls within Section 119 will be judged by reference to the state of the art as it existed on the date of the foreign application, rather than on the later date of the domestic filing. The difference is potentially crucial in the present case. During the interim between the filing of applications in Japan and the United States, the Patent Office granted Patent No. 3,831,929 (the “Hellmer patent”), which discloses all the components of the Rengo patent, and in the same arrangement. Rengo concedes that if the Hellmer patent is considered prior art, the Tokuno corrugator does not cross the non-obviousness threshold, and the ’502 patent must be adjudged invalid.
The parties agree that a patent application will receive the priority of Section 119 only if the foreign application satisfies the requirements of Section 112. Relevant to this case is the so-called “enablement” provision of Section 112: a patent application must “contain a written description of the invention, and of the manner and process of making and using it, ... as to enable any person skilled in the art to which it pertains ... to make and use the same.” Although the parties do not contest the point, this Court has never considered whether Section 119 incorporates the requirements of Section 112. Consequently, a brief examination of this issue must precede resolution of whether Rengo is entitled to the filing date of the Japanese application.
In the leading opinion on this subject, the Court of Custom and Patent Appeals decided that a foreign application must comply with the enablement provisions of Section 112 if the inventor is to enjoy the priority benefits of Section 119. Yasuko Kawai v. Metlesics, 480 F.2d 880 (C.C.P.A.1973). The court recognized that the language of Section 119 is not so unambiguous as to foreclose the contrary result. Because the statute provides that the American application, if preceded by a foreign filing “shall have the same effect as the same application would have if filed in this country” on the date of the earlier application, a question arises concerning the referent of the phrase “same application.” If it relates to the United States application, then the Ameri[548]*548can filing would be treated as though filed on the date of the foreign application, and only the American application need satisfy the requirements of Section 112. But if “same application” refers to the foreign claim, the foreign application must be examined for compliance with Section 112 before it is given an effective United States filing date.
Although resort to the statutory language is unavailing, the Section’s relationship with the body of American patent law assists the construction of Section 119. Section 104, 35 U.S.C. § 104, provides generally that an inventor cannot establish a date of invention by proving acts done in a foreign country, unless he falls within Section 119.26 With this one exception, the filing of the American application must be understood as a simultaneous conception and reduction to practice of the invention. Given the general policy embodied in Section 104, the Court in Kawai reasoned that the Section 119 exception “can aptly be described as a right to prove a date of invention no earlier than the date of the foreign filing.” 480 F.2d at 885. The foreign application, then, can be regarded as a reduction to practice, just as a domestic filing can be so regarded. Similarity in treatment is justified, however, only if similar safeguards are imposed: just as an American application will be regarded as a reduction to practice only if it describes the invention with the particularity required by Section 112, so must a foreign application satisfy this standard. To harmonize Section 119 with the general policy expressed in Section 104 requires that foreign applications be accorded priority only if they meet the enablement requirements of Section 112.
We find the Kawai syllogism persuasive.27 If a foreign application did not have to meet the requirements of American law, an inventor could file a foreign application for a promising but not yet perfected device, have one year to refine it before filing in the United States, and still avail himself of the priority grant of Section 119. We agree with the Court of Customs and Patent Appeals that the law should not be construed to tolerate this result.
Applied to the present case, the Kawai rule directs that Rengo may assert the Japanese filing date, and thereby exclude the Hellmer patent from the prior art, only if the Japanese application meets the disclosure standards of Section 112. The Japanese application is identical to the American in all relevant respects save one: it does not disclose circuit diagrams, included in the American applications, which explain how the action of the rotary shear is interrelated with the movement of the web diverter. Mr. Tokuno did not invent the specific circuitry disclosed in the ’502 patent, and did not claim it as part of the invention.
In considering the adequacy under Section 112 of the Japanese filing, the district court noted that the enabling requirement is directed, in the terms of the statute, to a “person skilled in the art,” and demands that the inventor describe the device in terms sufficient to enable one skilled in the art to make and use it. There is consensus in the present case that the appropriate level of skill in the art is very high, and is characterized by a person with an engineering degree who is thoroughly familiar with the design, construction, and operation of [549]*549corrugating equipment, and with the associated pneumatic, hydraulic, and electrical systems. After evaluating the expert testimony, the district judge concluded that the disclosures contained in the foreign applications would allow a person with the requisite skill “to make and use [the] invention without engaging in extensive experimentation and without requiring any inventive effort beyond the teachings of the Japanese documents.”
Molins contends that the district court misconstrued the disclosure requirements. According to Molins, a court may rely on skill in the art to interpret what is disclosed in the application, but the court may not speculate about what a skilled person could infer from what is disclosed. Molins does not dispute that a person sufficiently skilled could, after examining the Japanese application, infer that a particular type circuitry would be required for operation of the corrugator. Rather, it maintains that the failure to disclose, without more, precludes compliance with Section 112, thereby removing Rengo’s American application from the protective harbor of Section 119.
We find the Molins’ interpretation of Section 112 unduly parsimonious. It is axiomatic that no description, however detailed, is “complete” in a rigorous sense. Every description will rely to some extent on the reader’s knowledge of the terms, concepts, and depictions it embodies. Thus, an understanding of any description will involve some measure of inference. The Court of Customs and Patent Appeals recognized this principle in In re Wiggins, 488 F.2d 538, 543 (C.C.P.A.1973), when it said: “Every patent application and reference relies to some extent upon knowledge of persons skilled in the art to complement that disclosed in order that it be ‘enabling’ within the meaning of § 112.” See also In re Bode, 550 F.2d 656 (C.C.P.A.1977) (“One skilled in the art would be expected to have sufficient basic knowledge to construct such means, the structure of which would be simple.”). As Bode makes clear, skill in the art can be relied upon to supplement that which is disclosed as well as to interpret what is written.
The Bode interpretation of the enabling requirement has deep historical roots. Nearly one century ago, the Supreme Court, in Loom Co. v. Higgins, 105 U.S. 580, 585-86, 26 L.Ed. 1177 (1882), similarly construed the predecessor to Section 112:
If a mechanical engineer invents an improvement on any of the appendages of a steam-engine, he is not obliged, in order to make himself understood, to describe the engine, nor the particular appendage to which the improvement refers, nor its mode of connection with the principal machine. These are already familiar to others skilled in that kind of machinery. He may begin at the point where his invention begins, and describe what he has made that is new, and what it replaces of the old. That which is common and well known is as if it were written out in the patent and delineated in the drawings.
More recently other courts have expressed similar views. In Trio Process Corp. v. L. Goldstein’s Sons, Inc., 461 F.2d 66, 74 (3d Cir.), cert. denied, 409 U.S. 997, 93 S.Ct. 319, 34 L.Ed.2d 262 (1972), for example, we observed that “patent specifications need not be so detailed as production specifications,” and intimated that requirements of detail are less rigorous with respect to items that are not “a critical part of the invention,” id. at 74 n.24.
Applying these principles, we cannot say that the district court applied an incorrect standard of law. The circuit diagrams which Tokuno omitted from the Japanese application were not of his invention. Although the corrugator could not operate without circuitry, Bode and Higgins indicate that a patent need not specify those items which are not part of the invention, and which a person skilled in the art could be expected to understand. The district court, therefore, was correct in declining to find the Japanese documents inadequate for failing to disclose the circuit diagrams, and in evaluating them instead to determine whether a person with the requisite skill could infer from them a method for con[550]*550structing the machine. Moreover, we cannot say that the district court clearly erred in finding, based on first-hand observation of the evidence and testimony,28 that a person with the requisite skill who applied that knowledge to the apparatus disclosed in the Japanese application could have practiced the patent. The high level of sophistication which characterizes the person with skill in the art is thus, in this case, a double-edged sword. In the context of the Section 103 inquiry, the court might infer that a person with the requisite skill would find “obvious” many innovations which the layman would never imagine. But in the context of enablement, the knowledge of the skillful person, whose characteristics were enumerated in this lawsuit by Molins itself, facilitates a conclusion that the Japanese disclosure satisfies the American requirements. We will disturb neither the district court’s holding that the Japanese application satisfies the requirements of Section 112, nor the entailed holding, that Rengo is entitled, under Section 119, to the priority date of the Japanese filing.
Virtually inseparable from the enablement provisions are two additional prerequisites embodied in Section 112. The first paragraph of the Section states that “the specification shall contain a written description of the invention,” while the second paragraph stipulates that “the specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant reports as his invention.” Molins asserts that the '502 patent falls short of both these standards, known respectively as the “description" and “definiteness” requirements. Molins bases its argument on claim 1 of the ’502 patent, which recites, in pertinent part, a “means associated with the feeding guide plate for aligning said feeding guide plate with the lower guide plate, with the intermediate guide plate and with the upper guide plate.” This claim, Molins contends, requires a “means” — i. e., a diverting mechanism — that moves between three different positions. The specifications, however, disclose that the diverter moves between only two positions, one aligned with the upper flow path and the other with the lower flow path. Molins urges that this variance is fatal to Rengo’s asserted compliance with the description and definiteness elements of Section 112.
The requirements of adequate description and definite claim, though closely intertwined, are analytically distinct. The description requirement is contained in the same sentence as the enablement provision discussed above,29 and both direct inquiry to the person with ordinary skill in the art. The Court of Customs and Patent Appeals explained the scope of the requirement in Application of Smith, 481 F.2d 910, 914 (C.C.P.A.1973):
Compliance with the first paragraph of § 112 is adjudged from the perspective of the person skilled in the relevant art. This court has held that claimed subject [551]*551matter need not be described in haec verba in the specification in order for that specification to satisfy the description requirement . . . although where there is exact correspondence between the claim language and original specification disclosure, the description requirement would normally be satisfied.... The specification as originally filed must convey clearly to those skilled in the art the information that the applicant has invented the specific subject matter later claimed.... When the original specification accomplishes that, regardless of how it accomplishes it, the essential goal of the description requirement is realized.
The purpose of the description requirement is “to buttress the original filing date of the application as the prima facie date of invention,” 30 and the requirement typically is in issue when an applicant adds a claim after the original date of filing. Thus the Court in Smith observed: “Where the claim is an original claim, the underlying concept of insuring disclosure as of the filing date is satisfied, and the description requirement has likewise been held to be satisfied.” 481 F.2d at 914.
Description and enablement apply to specifications; by contrast, definiteness applies to claims. Definiteness means that “the language of the claims must clearly set forth the area over which the applicant seeks exclusive rights.” 31 The requirement essentially demands precision in the language of the claim, rather than adequacy of disclosure or description. Thus if the specifications are not commensurate with the subject matter encompassed by a claim, the claim may still be definite within the meaning of Section 112, although the specifications may not adequately describe the invention or enable one skilled in the art to reproduce it.32 The definiteness requirement is more than a linguistic quibble, however. Its purpose is to demarcate the boundaries of the purported invention, in order to provide notice to others of the limits “beyond which experimentation and invention are undertaken at the risk of infringement.”33 Thus, there is a subtle relationship between the policies underlying the description and definiteness requirements, as the two standards, while complementary, approach a similar problem from different directions. Adequate description of the invention guards against the inventor’s overreaching by insisting that he recount his invention in such detail that his future claims can be determined to be encompassed within his original creation. The definiteness requirement shapes the future conduct of persons other than the inventor, by insisting that they receive notice of the scope of the patented device.
In the case at hand, the requirement that the claim be definite presents no serious barrier to patent validity. The ’502 claim, which recites a means for aligning the feeder guide with the upper, middle, and lower flow paths, adequately notifies potential infringers of the scope and function of the claimed invention. Moreover, a relaxed standard of definiteness applies to combination patents. The final paragraph of Section 112 states:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material or acts described in the specification and equivalents.
The recitation in claim 1 of a “means” for aligning the feeding guide appears to comply with the standards of Section 112. While assertion of a “means,” without further specification, is admittedly somewhat nebulous, Section 112 contemplates that a reference to the “means” shall be understood, in connection with a combination pat[552]*552ent, to refer to a known method for performing the stated function identified in the specification. There remains the question whether the specifications correspond to the functions claimed, but as noted above, a lack of congruence will bear on disclosure, rather than on definiteness. Like the district court, we conclude that claim 1 of the ’502 patent satisfies the definiteness criterion of Section 112.
We also conclude that the specifications adequately describe the claimed invention. Molins emphasizes the asserted disparity between the recital of claim 1 and the specifications: while claim 1 recites a feeding plate that diverts the movement of the three-ply web and that can be aligned with three flow paths, the specifications describe a diverter that can be aligned with only two flow paths. We think that Molins has adopted an unnecessarily restrictive interpretation of claim 1. The patent claims a “means” for aligning the feeding plate with all three flow paths, but does not indicate that the feeder plate itself must move to each of the three positions. The specifications describe a system for aligning the plate with the middle flow path, even though there is no middle position on the feeding plate itself: the plate moves to the upper position, and a different mechanism, acting upon it, aligns it with the intermediate flow path. This system describes a “means” for aligning the feeder with all three flow paths, thereby removing the ground on which Molins objects to the adequacy of the description of the invention. See United States v. Adams, 383 U.S. 39, 49, 86 S.Ct. 708, 713, 15 L.Ed.2d 572 (1966).
V.
After finding the ’502 patent invalid, the district court, mindful of recommendations we have made in the past,34 proceeded to make findings on infringement. The court determined the accused machine to be equivalent in means, operation, and result to the Rengo corrugator, and further ruled that file-wrapper estoppel did not bar the plaintiffs from objecting to the Molins apparatus.35 The court accordingly held that the ’502 patent, if valid, was infringed.
In light of our disposition of the invalidity issue, we decline to review the district court’s rulings on infringement. As indicated above, the court’s invalidation of the ’502 patent must be vacated, but we will remand the case for reevaluation of the patent under the Graham guidelines. If the district court reinstates its ruling that the patent is invalid, and if this decision is upheld on appeal, then there will be no occasion to consider infringement. By declining to review the infringement issue now, we may thus avoid unnecessary resolution of a difficult and complex question. On the other hand, if the district court on remand upholds the ’502 patent, it will presumably, absent some intervening development, adhere to its decision on infringement. Then, on appeal, the infringement issue will properly be before us. But on the second appeal, if there is one, the infringement question will be no different than it is now, and no harm will accrue to the parties by our deferring consideration of the issue until that time when its resolution is necessary to dispose of the case. Moreover, deferral of the issue to a later time will delay final resolution of the dispute only on the assumption that Molins will decline to appeal a district court holding that the ’502 patent is valid. We have no basis for making this assumption. We believe prudence counsels in favor of declining to consider a question whose resolution either will prove unnecessary to a final disposition, or, if [553]*553necessary, will inevitably be before us again, none the worse for the delay.
VI.
The judgment of the district court will be vacated, and the case remanded for further proceedings in accordance with this opinion.