Dennison Mfg. Co. v. Ben Clements and Sons, Inc., 467 F. Supp. 391 (S.D.N.Y. 1979)
March 14, 1979
BEN CLEMENTS AND SONS, INC., Defendant.
United States District Court, S. D. New York.
*392 *393 *394 *395 *396 *397 James & Franklin, P. C., New York City, for plaintiff; Harold James, Robert L. Epstein, New York City, of counsel.
Blum, Moscovitz, Friedman & Kaplan, New York City, for defendant; Lawrence Rosenthal, William L. Botjer, New York City, of counsel.
STEWART, District Judge:
This is a suit for infringement of certain claims of a patent owned by plaintiff Dennison Manufacturing Company ("Dennison"), relating to clips of attachments as exemplified by the familiar molded plastic attachments used in retailing operations to affix tags to garments on display. The case was tried without a jury, and this opinion comprises the Court's findings of fact and conclusions of law pursuant to Rule 52(a), F.R.Civ.P.
Dennison is a corporation of the State of Nevada, with a place of business in Framingham, Massachusetts. It owns the patent in suit by virtue of an assignment by the patentee, Gordon B. Lankton ("Lankton").
Defendant Ben Clements & Sons, Inc. ("Clements") is a corporation of the State of New York, and at the time this suit was brought maintained a place of business in New York City. Clements imports and sells devices which allegedly infringe plaintiff's patent.
Jurisdiction and venue are not challenged. This is an action arising under the patent laws of the United States, 35 U.S.C. §§ 271 and 281, with jurisdiction based on 28 U.S.C. § 1338(a) and venue on 28 U.S.C. § 1400(b).
I. The Lankton Patent
The Lankton U.S. Patent 3,733,657, issued May 22, 1973 on an application filed May 25, 1972, concerns a plurality of attachments, each of which is to be associated individually with an object by being inserted through that object. The attachments are formed into an easily handled assembly by being secured together not only at those ends which are to be inserted through the object, as in the prior art, but also at their opposite ends, thereby preventing the individual attachments from tangling while they remain part of the assembly or clip. The second connection, that is, the one between the ends of the attachments that do not pierce the object, is designed to be more readily frangible in torsion than in tension, to facilitate the separation of an attachment from the clip after the attachment has been associated with an object, leaving the remaining attachments secured to one another.
Although the claims of the patent in suit claim more broadly, Lankton's original conception related to plastic molded attachments used in "marking" operations. In retail merchandising, it is often desirable to mark objects, such as garments displayed on a rack, to convey information to the consumer such as brand name, price and size. Over the years, many different means have been devised for and used in marking operations, but for the purpose of attaching a tag to a garment displayed on a rack, the most widely used device has been some *398 form of a "hang tag" which allows the tag to hang from the garment. At the time Lankton entered the picture in early 1972, Dennison was marketing (and still markets) a hang tag attachment known commercially as the "Swiftachment", which was introduced in 1965. It consisted of a number of plastic attachments integrally molded into a clip or assembly. Each attachment was comprised of a "T-bar" (the end which was inserted through an object by means of a hollow needle), an elongated section or "string", and at the opposite end of the string from the T-bar, an enlarged portion or "paddle", which was designed to prevent the tag once attached from falling off the string. Each attachment was joined at the T-bar end to a strip of plastic known as a "runner bar" by a sturdy connection which was designed to be cut by a blade within an attaching device, the entire assembly having been manufactured in a single molding operation. The paddle ends were not joined in any fashion. After being molded, the plastic assembly was stretched in order to impart strength and flexibility to the strings, a process which at first took place in the mold by moving the ends of the mold apart, and later, because of cost advantages, outside the mold in a separate stretching apparatus that grasped the ends of the attachments and pulled them in opposite directions. While out-of-the-mold stretching achieved production advantages over the previous method of in-the-mold stretching, it also presented a problem often when a clip was ejected from the mold one or more of its strings or paddles would deviate from the parallel orientation in which it was molded, with the result that the stretching apparatus would fail to grasp that paddle, leaving the string unstretched. Clips with just one unstretched string were not being shipped by manufacturers to customers at the time of Lankton's invention. Another, more serious problem, not particularly associated with out-of-the-mold stretching, was prevalent with the Swiftachment devices, as with its precursors the stretched strings had a tendency to tangle with other strings on the same clip and with the strings of other clips when packed in a box for shipment. It was to this problem that Lankton's invention specifically addressed itself.
The specification of the patent in suit relates several objects of Lankton's invention, which can be summarized as to provide a structural solution to the tangling problem. Analyzing the mechanics of a normal attaching operation utilizing the Swiftachment-type device, Lankton realized that if there existed a connection between the paddles of the attachments in a clip, that connection would be subjected to a combination of torsion and tension by the normal movements of the operator in attaching a tag to a garment. Thus, the structure Lankton devised was one that was designed to resist strong dislocative forces during manufacturing and handling, thus eliminating the tangling problem, but to break easily when twisted, in the course of an attaching operation. As indicated in the specification, Lankton thought the normal spacing between paddles, about 0.04 inch, militated against the attainment of a connection with the desired characteristics, so he devised a construction to shorten the distance, in effect making the connection more torsion-sensitive. Lankton provided protrusions on the face of each paddle surface, with the result that the filamentary connection spanned only 0.005 inch and consequently readily yielded to torsion.
Not all of the claims of the patent in suit, however, are limited to Lankton's conception. In addition to the protrusion/short filament embodiment, there are claims which cover generally clips of attachments joined at or near both ends by connections which are severed at different times and by different means, one connection being more readily severable than the other. As Clements' devices do not utilize the precise form of connection which Lankton conceived, the claims which defendant is accused of infringing are claims which are not limited to the protrusion/short filament embodiment.
II. The Issues
Defendant imports from Korea, Japan and Italy and sells in the United States *399 plastic molded assemblies of attachments utilizing one or more thin filamentary connections between the paddles. These are the alleged infringing devices. Clements denies infringement and asserts that the claims in suit are invalid because they are indefinite, because they "overclaim", and because they are obvious in light of the prior art. Even if the claims in suit are not invalid on any of these grounds, defendant argues, they may not be sued upon because they were obtained with deceptive intent. These are the issues to be decided in this opinion. In addition, defendant has filed a countersuit alleging patent misuse and antitrust violations, issues which have not yet been tried, and await the outcome of this suit. Damages are also reserved for later trial or accounting.
III. The Claims in Suit
The patent in suit contains 22 claims setting forth Lankton's invention and variations thereon, as well as methods of manipulating the assemblies of attachments. Defendant has not been accused of infringing all of these claims, and there is a threshold question as to which claims are in suit.
The Pre-Trial Order in this case listed as an issue to be tried whether the following claims are infringed by defendant: Claims 1, 2, 3, 4, 5, 7, 9, 10, 11, 14, 15, 16 and 17. However, in its Trial Brief and again in its Post-Trial Brief, plaintiff stated that it was confining its case "without prejudice" to proving infringement of Claims 1, 2, 4, 5, 7, 10, 11, 16 and 17, thus omitting Claims 3, 9, 14 and 15 from its proof. We therefore will make no determination of whether Claims 3, 9, 14 and 15 are infringed. But that does not end the matter. Defendant asserted in its answer to the complaint in this case the affirmative defense of invalidity of the claims in suit. Under Rule 8(c), F.R.Civ.P., we are required to treat a mistakenly designated defense as a counterclaim where justice so dictates. Applying this standard, and treating the affirmative defense of the claims' invalidity as a properly designated counterclaim, we think defendant is entitled to an adjudication of whether all claims which it has been accused of infringing are valid, including those which plaintiff has chosen to omit from its proof. Since defendant arguably does infringe the omitted claims, there exists a genuine controversy as to the validity of the claims and that issue ought to be decided. Thus, we will consider the validity of all of the claims denominated in the Pre-Trial Order, which are set forth in full in the margin.
*400 Claims 1, 2, 3, 4, 5, 7, 9, 10 and 11 are article claims, describing the structure claimed in the patent in suit. Claim 1 is the independent claim upon which the remainder of the claims directly or indirectly depend, that is, they incorporate each and every limitation of Claim 1 and add further requirements or refinements. Claims 14, 15, 16 and 17 are method claims, claiming as invention the methods of manipulating the assemblies of attachments.
We consider first the validity of the claims in suit and then their infringement.
IV. Presumption of Validity
The validity of the claims in suit is strenuously challenged by defendant if the claims in suit are invalid, they cannot be infringed. Clements asserts their invalidity on three grounds: indefiniteness (35 U.S.C. § 112), "overclaiming" (35 U.S.C. § 112), and obviousness (35 U.S.C. § 103). There is no attack on the novelty (35 U.S.C. § 102) of the claimed invention.
35 U.S.C. § 282 reads in relevant part:
A patent shall be presumed valid. Each claim of a patent (whether in independent, dependent, or multiple dependent form) shall be presumed valid independently of the validity of other claims: dependent or multiple dependent claims shall be presumed valid even though dependent upon an invalid claim. The burden of establishing invalidity of a patent or any claim thereof shall rest on the party asserting such invalidity.
This statutory presumption of validity has no evidentiary weight but means that reasonable doubts will be resolved in favor of the patentee. Reeves Bros., Inc. v. U. S. Laminating Corp., 417 F.2d 869, 872 (2d Cir. 1969); Lorenz v. F. W. Woolworth Co., 305 F.2d 102, 105 (2d Cir. 1962); Lerner v. Child Guidance Products, Inc., 406 F. Supp. 560, 563-64 (S.D.N.Y.1975), aff'd, 547 F.2d 29 (2d Cir. 1976). Its effect on the burden of proof is that the patent holder is relieved of the burden of establishing the validity of the claims in suit as a requisite for the successful maintenance of an infringement action and the burden is placed instead on the alleged infringer who asserts the invalidity of the claims. Lorenz v. F. W. Woolworth Co., supra, 305 F.2d at 105. A preponderance of the evidence will satisfy the burden. Plastic Contact Lens Co. v. W. R. S. Contact Lens Labs, Inc., 330 F. Supp. 441, 443 (S.D.N.Y.1970); Thomson Industries, Inc. v. Nippon Thompson Co., 298 F. Supp. 466, 472 (E.D.N.Y.1968).
These principles apply generally to our discussion of the validity of the claims in suit, to which we now turn. However, the presumption of validity will require further *401 analysis in the context of defendant's argument that the claims are invalid because of obviousness, infra.
V. 35 U.S.C. § 112
Defendant argues two grounds on which the claims in suit are invalid under 35 U.S.C. § 112: indefiniteness and "overclaiming". That section reads in pertinent part:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Defendant asserts that the claims in suit fail to satisfy the specificity requirement of § 112 in three ways. It is argued first that the claims contain impermissible "terms of relativity"; second, that the terms of the claims are vague and indefinite; and third, that the claims do not define the nature of the inner-paddle connection or the method of severing it with sufficient specificity. Plaintiff responds that the claims are clear and definite and that defendant distorts the claim language. Plaintiff urges the conclusion that the claim language is meaningful and understandable to those skilled in the art.
To be patentable "[a]n invention must be capable of accurate definition, and it must be accurately defined . . .." United Carbon Co. v. Binney & Smith Co., 317 U.S. 228, 237, 63 S. Ct. 165, 170, 87 L. Ed. 232 (1942). The specificity required by § 112 advances two objectives:
[F]irst, that in return for the grant of a patent monopoly the patentee must make sufficient disclosure to enable one skilled in the art to reconstruct the invention; and, second, that the patentee describe his invention with sufficient definiteness to enable others to discern the boundaries beyond which experimentation and invention are undertaken at the risk of infringement.
Norton Co. v. Bendix Corp., 449 F.2d 553, 555 (2d Cir. 1971). In consonance with these objectives, absolute precision in the claim language is not required; it is enough "[i]f the claims, read in the light of the specifications, reasonably apprise those skilled in the art both of the utilization and scope of the invention, and if the language is as precise as the subject matter permits . . .." Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124, 136 (2d Cir.), cert. denied, 358 U.S. 884, 79 S. Ct. 124, 3 L. Ed. 2d 112 (1958).
Defendant argues, incorrectly, that we may not resort to the specification to determine the meaning of terms in the claims. "[I]t is fundamental that claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention . . .." United States v. Adams, 383 U.S. 39, 49, 86 S. Ct. 708, 713, 15 L. Ed. 2d 572 (1966). While the specification may not expand or limit the patent monopoly, which is measured by the claims, Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., 340 U.S. 147, 149, 71 S. Ct. 127, 95 L. Ed. 162 (1950), rehearing denied, 340 U.S. 918, 71 S. Ct. 349, 95 L. Ed. 663 (1951), it is entirely appropriate to resort to the specification for the purpose of understanding patent claims that are complex, unclear, or ambiguous. Maclaren v. B-I-W Group Inc., 535 F.2d 1367, 1372-73 (2d Cir.), cert. denied, 429 U.S. 1001, 97 S. Ct. 531, 50 L. Ed. 2d 612 (1976). With these principles in mind, we turn to defendant's arguments that the claims are invalid for lack of specificity.
*402 First, Clements contends that the claims in suit utilize "terms of relativity" which are impermissible in that they evade precise definition. The terms which defendant cites as objectionable are found in Claim 1 ("comparatively more readily severable"), Claims 3 and 9 ("more readily severable in torsion than in tension"), and Claims 4 and 10 ("substantially greater resistance to breaking when tensioned than when twisted"). Defendant relies on Long Mfg. Co. v. Lilliston Implement Co., 328 F. Supp. 268, 276 (E.D.N.C.1971), aff'd per curiam, 457 F.2d 1317 (4th Cir.), cert. denied, 409 U.S. 874, 93 S. Ct. 119, 34 L. Ed. 2d 126 (1972), in which the court held invalid for indefiniteness a patent on a threshing machine because it used terms such as "relatively large" when describing one of the threshing cylinders and "relatively low" when referring to the height of the combine, without supplying in the claims or specification any dimensions or bases for comparison. The claims before us, however, are markedly different from those in Long, in that bases for comparison are provided within the language of the claims themselves. In Claim 1, the severability of the second connection, the "securing means", is compared to that of the "frangible means" joining the T-bar end to the runner bar. In Claims 3 and 9, the characteristics of the connection under tension are compared to its characteristics when subjected to torsional forces; and the effect on the paddle connection of tension and twisting forces are contrasted in Claims 4 and 10. We have no difficulty understanding these claims, and they should not present a problem to one skilled in the art, particularly when read in the light of the specification and drawings.
Next, defendant argues that the word "twisted", found in Claims 4 and 10, is vague and indefinite, as evidenced by the fact that its meaning was disputed by defendant's and plaintiff's experts. Defendant adduced testimony that "twist", as used in these claims, meant rotation in a single plane and that the connections claimed in the patent in suit were intended to break upon the application of rotation alone; or, alternatively, that they were intended to be broken by the sequential rather than simultaneous application of torsion and tension. Plaintiff argued that Lankton's meaning of "twist" included a slight pull or a corkscrew motion. The existence of a conflict over the possible meaning of terms in a claim does not require the conclusion that the claim is indefinite. We think it clear when the claims, specifications and drawings are read together that "twist" was meant to and does include a pulling motion in addition to rotation.
Defendant challenges the specificity of the language in Claims 1, 14 and 16 that the paddle connection is designed to be and is broken "externally of an attaching device." With respect to Claim 1, when this language is read in its context and in light of the specification and drawings, it is evident that the paddle connection is intended to be severed while the clip of attachments remains in the gun. The "comparatively more readily severable" connection is broken as part of a normal attaching operation, in which the leading attachment is "twisted off from the clip." Col. 4, line 32; see Figures 3 and 4. No separate action is required on the part of the operator, other than the "continuous movement which is part of a normal attaching operation." Col. 3, lines 42-46; see also Col. 6, lines 15-17. Thus, in the context of Claim 1, rather than meaning that the connections between the paddles are severed after the clip is removed from the gun, "externally of said attaching device" is more appropriately interpreted as contrasting the means of severing the paddle connection to the means of severing the T-bar connection, which is severed in and "by means of an attaching device." With respect to Claims 14 and 16, plaintiff concedes that the "externally of an attaching device" language found in these claims yields the meaning ascribed to it in Claim 1 as well as the additional meaning urged by defendant that the connections *403 are severed after the clip is removed from the gun, as in the "batch" embodiment described in Col. 8, lines 26-36. We think that Claims 14 and 16 may be read to describe both embodiments, and that the significance and meaning of "externally of an attaching device" is not thereby diminished. In both embodiments, the T-bar connection is broken by means of a blade in the attaching device which is activated when the operator "pulls the trigger" of the gun. Immediately after the T-bar connection has been severed, the T-bar is pushed through the needle of the attaching device, attaching the leading attachment to an object. In the case of the embodiment of Figures 1-13, the paddle connection is then severed by the operator's pulling the gun, still containing the clip of attachments, away from the leading attachment. It is severed "externally" of the gun in that, unlike the T-bar connection, no blade or other mechanism within the attaching device effects the separation, which occurs outside of the device. This meaning is clearly set forth in the specification and drawings. In different fashion, but nonetheless externally of the attaching device, the paddle connection in the batch embodiment is severed by a knife or scissors, after the clip has been removed from the gun. This meaning, too, is clearly set forth in the specification. Thus, the claim language that the connection is broken "externally of an attaching device" is not indefinite or insufficiently specific. That the language reads broadly does not serve to invalidate the claims so long as its meanings are clear in light of the specification and drawings, as they are here.
In addition to attacking the specificity of particular phrases of the claims, defendant assails the article and method claims in suit for failing to define the nature of the paddle connection and the method of severing it. This defect in the broader claims, such as Claims 1, 14 and 16, defendant argues, is not cured by the narrower claims, which either simply locate the connection or are themselves vague and indefinite, and thus must also be adjudged invalid. Claim 1 defines the "securing means", or paddle connection, as "being comparatively more readily severable than [the T-bar connection] and adapted to be severed externally of an attaching device." Defendant argues that the "comparatively more readily severable" portion of this definition is vague and indefinite, so much so that a paddle connection which is a miniscule degree smaller and weaker than the T-bar connection and which cannot be severed except by means of a knife or scissors would fall within the definition. This interpretation ignores the limitation contained in the claim that the paddle connection be "adapted to be severed externally of" the attaching device. This limitation, which was added to the claim by amendment after rejection because of obviousness, when read in the light of the specification and drawings, excludes from the scope of the claim defendant's suggested interpretation. The references in the specification which describe how the connection defined in Claim *404 1 is to be broken best illustrate the meaning of the limitation. The leading attachment is separated by being twisted off from the remaining attachments in the clip. This twisting force is applied as part of a normal attaching operation, requiring no separate action on the part of the operator. Col. 2, lines 49-52, 65; Col. 3, lines 2, 13-17, 40-45.
. . . [E]ach attachment as it is individually used may readily be separated from the clip without requiring any separate action on the part of the operator other than a continuous movement which is part of a normal attaching operation in any event.
Col. 3, lines 40-45. Thus, the connection is broken by the motion of pulling the attaching gun away from the leading attachment which has been fastened to an object (See Figures 3 and 4 and their description in the specification, Col. 4, lines 25-34); and the clear meaning of the "adapted to be severed externally" language is that the structure of the paddle connection is such that it can be broken as described in the specification. It would be inconsistent with this language as clarified by the specification to say that it includes a connection that is to be severed by a knife or scissors. Defendant's argument that Claim 1 fails adequately to define the paddle connection is an exercise in selective reading of the claim. Maurice A. Crews, an expert witness produced by defendant, expressed his opinion that the claim was vague as to the paddle connection, but failed to read the claim in its entirety. The claim states that the paddle connection is "comparatively more readily severable than [the T-bar connection] and adapted to be severed externally of an attaching device" (emphasis added), but Crews refused to read these phrases of the claim together, let alone read them with the specification and drawings. Tr. at 2771-72, 2823, 3111-12. We conclude that defendant has not met its burden of proving that Claim 1 is invalid for failure to define the paddle connection. It is defined sufficiently well to apprise one skilled in the art of the scope and utilization of the invention, and that is all that is required. Georgia-Pacific Corp. v. United States Plywood Corp., supra, 258 F.2d at 136.
The argument is made that while Claims 2, 5, 7 and 11 specify the location of the second connection, they are of no assistance in further defining the connection. Given our finding that Claim 1, when read in the light of the specification and drawings, defines the nature of the connection with sufficient specificity to satisfy the requirements of 35 U.S.C. § 112, we find nothing objectionable in Claims 2, 5, 7 and 11.
Claims 3 and 9 define the second connection as being "more readily severable in torsion than in tension". Claim 3 is dependent upon Claim 2, which places the connection defined in Claim 1 "between and connected to" the paddles, while Claim 9 depends directly upon Claim 1, and thus does not contain the Claim 2 limitation. Defendant argues that the meaning of "more readily severable in torsion then in tension" is "mysterious", Tr. 2780, vague and indefinite. Claims 3 and 9, on their face, are not definitive as to the nature of the connection, but resort to the specification yields a clear understanding of what is required of the connection and how it is intended to function. It is an object of the invention to design a connection that would remain intact during manufacturing and handling, where it would be subjected to tension and shear forces, but that would break easily as part of the normal attaching operation, where it would be twisted off from the clip, that is, subjected to torsional forces. Col. 2, lines 39-48. The significance of the differentiation between torsion and tension is, thus, apparent in the specification, *405 and these claims adequately detail the relationship between the two.
Claims 4 and 10 define the paddle connection as comprising "a short, narrow length of material and dimensions such as to have a substantially greater resistance to breaking when tensioned than when twisted." Claim 4 is dependent upon and incorporates the limitation of Claim 2, while Claim 10 depends directly upon Claim 1. Defendant admits that Claims 4 and 10 further define the nature of the connection, but contends that they nevertheless are vague and indefinite. Tr. 2337, 2782-84. We disagree. Although the claims do not specify how short and narrow the connection must be, the specification defines the properties of the connection, describes how it is to perform, and even sets forth Lankton's "best mode", 35 U.S.C. § 112, giving the exact dimensions of the connection favored by Lankton. When read in the light of the specification and drawings, the meaning of "short" and "narrow" would be readily apparent to one skilled in the art.
Thus, we find that the article claims in suit are not invalid for failure to particularly point out and distinctly claim the subject matter of the invention. We turn our attention next to the method claims and consider whether each of these meets the specificity requirements of 35 U.S.C. § 112.
Defendant attacks Claims 14 and 16 on the grounds that they "merely recite the sequence of severing and that the second connection is severed outside of the attaching device. Undefined are the nature of the connection and the method of severing same, essential limitations in these claims." Defendant's Post-Trial Brief, 16. That the method claims fail to define the nature of the connection is no shortcoming. A process or method is patentable in and of itself, 35 U.S.C. §§ 100(b) and 101, there being no requirement that a structure be disclosed in a method claim. As to the argument that Claims 14 and 16 are unspecific in that they fail to specify when and how the paddle connection is severed, resort to the specification provides the necessary definiteness. As we noted above, both the device of Figures 1-13 and that of Col. 8, lines 26-36, the batch embodiment, are encompassed by the language of Claims 14 and 16. And, as we also noted above, the fact that these claims can be interpreted to encompass both devices and methods of severing their connections does not invalidate the claims, as long as both are defined with the requisite specificity, as they are here.
Nor are Claims 15 and 17 indefinite. Defendant argues that they suffer the same infirmity as Claims 14 and 16, the claims upon which they respectively depend; that is, that they do not specify how or when the second connection is to be broken and include within their scope the batch embodiment. We disagree that these claims encompass the batch embodiment. Claims 15 and 17 accomplish the separation of the leading attachment from the clip "by rotating said attachment relative to said clip, thereby . . . exert[ing] a substantial torsional force" on the paddle connection. In the batch embodiment, the paddle connection is "highly strain-resistant" and is broken "by cutting it with a knife or scissors." Col. 8, lines 26-36. Thus Claims 15 and 17 do not read on the batch embodiment but are confined to the devices depicted in Figures 1-13, in which the leading attachment is separated from the clip by rotating it relative to the clip as part of a normal attaching operation.
For the foregoing reasons, we conclude that none of the claims in suit are invalid for failure to point out and distinctly claim the subject matter of Lankton's invention.
Because the claims in suit omit the critical limitations required to define Lankton's invention, defendant argues, they encompass not only his invention but vast areas of the art not properly a part of the patentee's monopoly. These omissions, it is argued, invalidate the claims because they fail to comply with the mandate of 35 U.S.C. § 112 that:
*406 The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Since Lankton's conception was limited to the short filament/protrusion embodiment of Col. 6, lines 18-36, defendant contends that the claims in suit, none of which are so limited, are invalid because they do not claim "the subject matter which the applicant regards as his invention."
None of the authorities cited by defendant supports this argument. Overclaiming is properly asserted as a defense when the claims include by their breadth "that which is conceded to have been known in the prior art that which the prior art anticipated or made obvious." Wayne Knitting Mills v. Russell Hosiery Mills, Inc., 400 F.2d 964, 968 (4th Cir. 1968), cert. denied, 393 U.S. 1064, 89 S. Ct. 717, 21 L. Ed. 2d 707 (1969); accord, Wilcox Mfg. Co. v. Eastern Gas & Fuel Associates, 400 F.2d 960, 964 (4th Cir. 1968), cert. denied, 393 U.S. 1051, 89 S. Ct. 691, 21 L. Ed. 2d 693 (1969); see Jamesbury Corp. v. Litton Industrial Products, Inc., 586 F.2d 917 (2d Cir. 1978). Defendant does not argue that the claims in suit were anticipated by the prior art, and the issue of obviousness is one which we shall discuss in great detail, infra. It has also been recognized as a defense where the claims were not limited to a structure essential to accomplish the advantage claimed for the invention. Murton v. Ladd, 122 U.S.App.D.C. 274, 276, 352 F.2d 942, 944 (1965). But that is not the situation here, for we find that the claims in suit disclose and are limited to a structure which is essential to carry out the advantages set forth in the specification and drawings. That they vary from Lankton's actual conception of the structure necessary to accomplish the results he envisioned does not detract from their validity. Claims will almost always be broader than the inventor's own conception as revealed in the specification and drawings, since he is only required to set forth the "best mode", 35 U.S.C. § 112, of carrying out the invention. American Photocopy Equipment Co. v. Rovico, Inc., 257 F. Supp. 192, 196 (N.D.Ill.1966), aff'd, 384 F.2d 813 (7th Cir. 1967), cert. denied, 390 U.S. 945, 88 S. Ct. 1030, 19 L. Ed. 2d 1133 (1968). The breadth of the claims does not itself cause their invalidity it merely enhances the risk that the claims will be anticipated or made obvious by the prior art. Id.
The pivotal issue in this lawsuit is that the claims in suit are invalid because they are obvious in light of the prior art. Before turning to an analysis of the prior art and its teachings, it is appropriate to set forth as background some principles of law which pertain to that analysis.
As we stated above, a patent is presumed valid, and the burden of proof rests on one who asserts its invalidity. 35 U.S.C. § 282; e. g., Lorenz v. F. W. Woolworth Co., supra, 305 F.2d at 105. However, the patent in suit, being a combination of old and known elements relying for its patentability on the assertedly new and different function produced by the combination, should be subjected to more careful scrutiny than a patent which introduces novel elements into the art.
Courts should scrutinize combination patent claims with a care proportioned to the difficulty and improbability of finding invention in an assembly of old elements. The function of a patent is to add to the sum of useful knowledge. Patents cannot be sustained when, on the contrary, their effect is to subtract from former resources freely available to skilled artisans. 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.
Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., supra, 340 U.S. at 152-53, 71 S. Ct. at 130; accord, Sakraida v. Ag Pro, Inc., 425 U.S. 273, 281, 96 S. Ct. 1532, 47 L. Ed. 2d 784 (1976); Anderson's-Black *407 Rock, Inc. v. Pavement Salvage Co., Inc., 396 U.S. 57, 60-61, 90 S. Ct. 305, 24 L. Ed. 2d 258 (1969); Julie Research Laboratories, Inc. v. Guildline Instruments, Inc., 501 F.2d 1131, 1136 (2d Cir. 1974).
In addition to this caveat, defendant urges that the presumption of validity of the claims in suit is weakened or negated by the fact that the Patent Office did not have before it all the pertinent prior art when it examined Lankton's application. While it is true that the failure on the part of an applicant to bring to the attention of the Patent Office pertinent prior art may weaken the presumption of validity of the claims, to have that effect the omitted prior art must be more than merely cumulative. Julie Research Laboratories, Inc. v. Guildline Instruments, Inc., supra, 501 F.2d at 1136 ("significant" prior art not before patent examiner); Reeves Brothers, Inc. v. U. S. Laminating Corp., supra, 417 F.2d at 871-72 ("certain relevant prior patents", which essentially disclosed applicant's invention, were not cited to Patent Office); Lerner v. Child Guidance Products, Inc., supra, 406 F. Supp. at 563 (strength of the presumption depends on whether the Patent Office had before it "the most relevant prior art"); Kahn v. Dynamics Corporation of America, 367 F. Supp. 63, 67 (S.D.N.Y. 1973), aff'd, 508 F.2d 939 (2d Cir.), cert. denied, 421 U.S. 930, 95 S. Ct. 1657, 44 L. Ed. 2d 88 (1975) ("applicable prior art" was not before the Patent Office, as applicant had misrepresented the teachings of a prior patent over which applicant's invention had to define in order to be patentable). Here, as will be evident from our discussion of the prior art, the patent examiner had before him prior art no less relevant to the claims of the patent in suit than the references cited to us by defendant. The presumption of validity, therefore, remains intact and is, in fact, strengthened.Lerner v. Child Guidance Products, Inc., supra, 406 F. Supp. at 563.
A. The "John Deere" Inquiry
Although the ultimate question of patent validity is one of law, Graham v. John Deere Co., 383 U.S. 1, 17, 86 S. Ct. 684, 15 L. Ed. 2d 545 (1966), the obviousness or non-obviousness of the subject matter is to be determined against a factual background.
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 resolved.
Id. As evidence of obviousness, defendant presented prior patents as well as the testimony of its two expert witnesses. The first, Harvey W. Samo, a mechanical engineer with some study in and a great deal of practical experience with plastic molding technology, testified to the obviousness of Lankton's invention in light of the prior patents and the plastic molding art in general. Defendant's second expert was Maurice A. Crews, a "largely retired" patent attorney who had been employed at one time by the Patent Office and who, by his own admission, had practically no experience *408 in the marking field, Tr. at 2693, no experience in plastic molding, Tr. at 2694, and no contact other than in connection with this lawsuit with the guns and attachments in issue. Tr. at 2694. Mr. Crews' particular expertise with respect to this lawsuit consisted almost entirely of that garnered from his study of the patent in suit and of the prior art references pertinent to this litigation. Tr. at 2694. A large portion of his testimony was an interpretation of the claims of the patent in suit in connection with defendant's indefiniteness argument, and, despite his lack of familiarity with the plastic molding and attaching arts, a discussion of modifications of the prior patents that might be made by one skilled in the art. Plaintiff's expert, Michael Rabins, a mechanical engineer and chairman of the department of mechanical engineering at Wayne State University, possessed academic expertise in the fields of statics, strength of materials, kinematics and machine design. His qualifications also included experience as a consultant and expert witness in a previous trial involving the attaching field, and consulting work for toy companies involving plastic, plastic molding and the behavior of plastic parts. Tr. 1325-26. Professor Rabins' testimony was concerned mostly with the infringement issue, but he also commented on the prior art. On the question of obviousness, therefore, the most relevant, credible and significant evidence is the prior art publications, and the testimony of Samo and, to a lesser extent because of his lack of specialized knowledge of the plastic molding and attaching arts, Rabins.
Scope and Content of the Prior Art
The parties dispute the scope and content of the pertinent prior art. Plaintiff has taken somewhat inconsistent positions with respect to the prior art against which the claims of the patent in suit are to be measured. In prosecuting this patent, the applicant took a limited view of the scope of the prior art. Each and every reference cited to the patent examiner was drawn from the tag or tag-attaching art (the Schumm patent, infra, was independently cited by the examiner). In its Trial Brief, at 69, plaintiff suggests that the relevant prior art "is that of molded plastic attaching devices of the type exemplified by Super Swiftachments," a narrow view but one claimed to be "strictly commensurate with the claim language". In its Post-Trial Brief, at 53, plaintiff repeats its assertion that its view of the relevant prior art is "strictly commensurate with the claim language" but fails to say what its view is of the prior art. Finally, in its Reply Brief, at 24, plaintiff seemingly retreats from its earlier position and states simply that "a flexible plastic special purpose attachment is not a rigid nail," arguing that the Keck patent, infra, should not be considered prior art because of the different considerations applicable to rigid nails as opposed to flexible plastic attachments. Defendant has consistently argued that the pertinent prior art should be "the plastic molding art and the attaching or fastening art." Post-Trial Brief, 25. For the reasons that follow, we agree with defendant's position.
In Digitronics Corp. v. New York Racing Ass'n, Inc., 553 F.2d 740 (2d Cir.), cert. denied, 434 U.S. 860, 98 S. Ct. 187, 54 L. Ed. 2d 133 (1977), the court considered as relevant prior art the data processing art in general rather than data processing as applied to the totalisator art, relying on Dann v. Johnston, 425 U.S. 219, 96 S. Ct. 1393, 47 L. Ed. 2d 692 (1976) (a case very similar to Digitronics on its facts Court applied as prior art computer technology generally rather than computer technology in the banking industry). The court stated:
In determining the scope of the relevant prior art with which the hypothetical ordinarily skilled person must be presumed to be familiar, we hold simply that the court must look, in light of both the training of the patentee and the elements in the claimed invention which give it its *409 novel quality, at what arts the patentee could reasonably be expected to consult in doing the inventing.
Digitronics Corp. v. New York Racing Ass'n, Inc., supra, 553 F.2d at 745. The first consideration, the training of the patentee, presents no difficulty. Lankton's academic training was as a mechanical engineer and his professional experience up to the time of his testimony in this trial consisted of twenty or more years in the plastic molding field, involving the manufacture of a wide variety of products. Tr. at 892-93. Moreover, his familiarity with the plastic molding field covers the full range of thermoplastic materials. Tr. at 894. Therefore, Lankton could reasonably be expected to consult the entire plastic molding art in devising his invention.
The second consideration, the elements of the invention which give it its claimed novel quality, requires that we also look to the claims to determine the scope of the pertinent prior art. It is the "claims which define the boundaries of a patent monopoly." Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., supra, 340 U.S. at 149, 71 S. Ct. at 129; Maclaren v. B-I-W Group Inc., supra, 535 F.2d at 1372. Since by the granting of a patent a patentee is granted a monopoly as broad as his claims may reach, "[i]t is but an evenhanded application to require that those persons granted the benefit of a patent monopoly be charged with an awareness", Deere, supra, 383 U.S. at 19, 86 S. Ct. at 695, of the applicable art. The breadth of the claims, therefore, in part determines the scope of the prior art. Maclaren v. B-I-W Group Inc., supra, 535 F.2d at 1376. Claim 1 of the patent in suit is not limited to "molded plastic attaching devices of the type exemplified by Super Swiftachments," or to "flexible plastic special purpose attachment[s]," definitions of the relevant prior art alternatively offered by plaintiff, but is a broad claim describing "a clip of a plurality of attached-together attachments." This claim is easily broad enough to include within its scope the Keck device whose claims describe "an assemblage of collated fasteners" as well as many other devices in the attaching or fastening art. The same may be said of the remainder of the article claims in suit, which depend directly or indirectly upon Claim 1, and of the method claims in suit. Further, although the specification refers to attaching devices molded of thermoplastic material, none of the claims in suit is limited to molded plastic attaching devices or to flexible plastic special purpose attachments; rather, each claim relates to attachments or clips of attachments in general without specifying the materials from which the attachments are to be fashioned. The breadth of the claims appears to be a result not of oversight, but of intention. While Lankton's application as originally submitted for Patent Office consideration contained dependent claims (Claims 14-20) which specified clips "constituted by a one-piece assembly of molded plastic material," Defendant's Exhibit A, the amended claims omit any reference to materials. From this, it seems evident that the claims in suit, which are not so limited, were intended from the start to encompass within their scope "clip[s] of a plurality of attached-together attachments," regardless of their material composition. Therefore, plaintiff may not now limit the relevant prior art to plastic molded attaching devices. See Maclaren v. B-I-W Group Inc., supra, 535 F.2d at 1373. Rather, the scope of the relevant art, as defined by the claims in suit, should include, as defendant suggests, the attaching or fastening art in general. See Deere, supra, 383 U.S. at 19, 86 S. Ct. 684.
Having determined that the relevant prior arts include both the plastic molding art and the attaching or fastening arts, we now turn to the question of the content of those arts as they relate to the claims in suit.
In determining the content of the prior arts which are relevant to the claims in suit, we consider those prior art references cited by the patent examiner in granting the patent in suit, those additional references cited by plaintiff in the application which were not cited by the examiner, *410 and the patents and prior art cited by defendant which were not before the examiner. In prosecuting its application for the patent in suit, plaintiff applied for and was granted special status, subjecting the application to accelerated examination. In fulfilment of the applicant's obligation under the accelerated examination procedure to make a detailed disclosure of prior art references, plaintiff listed eleven prior patents. In rejecting as obvious nineteen of the twenty-four claims of plaintiff's application (and objecting to the remaining five as dependent upon the obvious claims), the patent examiner cited a twelfth prior patent. We begin by analyzing the disclosures contained in these twelve patents and then proceed to examine the eight prior art references (seven U.S. patents and one Japanese Utility Model Publication) cited by defendant that were not brought to the attention of or cited by the patent examiner.
First, we consider the three patents cited by the patent examiner in allowing the claims of the patent in suit: Bone 3,103,666, Schumm 3,147,522 and Flood 2,825,162. Bone 3,103,666, patented September 17, 1963 and assigned to Dennison, is entitled Tag Attaching Apparatus and relates to "bar-lock" tag attachments. The bar-lock attachments with which Bone's apparatus, or gun, is to be used are integrally molded in a strip of plastic material, each bar-lock consisting of a bar, a button and a filament joining the two. The bars are joined to a runner bar, forming a clip which is fed through Bone's gun. The specification of the Bone patent mentions that previous attachments, namely the "bar-string-button" devices, with their string filaments, were susceptible to tangling and could not conveniently be attached by use of an automatic device. The improved plastic-molded bar-lock device lent itself easily to use in an automatic attaching device, the subject of Bone's patent, and impliedly eliminated the tangling problem. Schumm 3,147,522, patented September 8, 1964 and entitled Flexible Tie, discloses a tightening device designed to be drawn in the form of a loop about an article such as the open end of a bag. The devices are molded parallel to one another and "connected by a plurality of narrow connecting portions integral with the enlarged end portions, which connecting portions may be easily severed to provide thus for a plurality of separate strips." The advantage obtained by connecting the devices one to another is that the clip of fasteners may be fed into "a packing machine in which the connecting portions . . may be severed and in which each of the thus severed strips may automatically be formed into a loop and tightened about an end portion of a filled bag to close the same. . . ." Thus, Schumm teaches a clip of fasteners, each of which is joined to its adjacent fasteners at each end by narrow connecting portions, which results in its suitability for use in conjunction with an automatic packaging machine. Both connections are severed in the machine. Flood 2,825,162, patented March 4, 1958, assigned to Dennison and denominated String Tag Attachment Device, discloses a loop of string with a bar receiving the ends of the loop and a button affixed at the loop's midpoint. "The devices are grouped together in a row with the bars in parallel juxtaposition and a strip of material extending along the row, the strip being temporarily joined to each device of the group with adhesive which holds the group together but which permits the devices to be detached individually." The strip, whose purpose as disclosed in the patent is to permit convenient dispensing, preferably is joined to the string of each device, but may instead be secured to the bars. The patent teaches that the bar of the leading attachment may be placed in the needle before the leading attachment is peeled away from the adhesive strip.
In addition to these patents which were cited by the examiner, the specification of the patent in suit refers to two prior art patents: Bone 3,444,597 and Kirk 3,380,122. Bone 3,444,597, patented May 20, 1969, assigned to Dennison and entitled Filament Type Attachment, teaches a one-piece, molded plastic attachment device which claims to improve upon the devices to which Bone 3,103,666, relates (the molded plastic *411 bar-lock devices) by stretching the portion which connects the bar and the button (the string), thus adding flexibility and strength to the string portion. In its preferred form, ". . . a plurality of the devices are molded together, the devices being connected by a rod . . . integrally joined to the cross-bars by necks . . .." Bone 3,444,597 discloses the Dennison Swiftachment device, a clip of attachments with stretched strings, each attachment being connected at one point to a runner bar by a neck that is designed to be severed in a machine or apparatus. The patent was disclaimed and dedicated to the public on October 24, 1970. Kirk 3,380,122, also assigned to Dennison, was patented April 30, 1968. Entitled Mold for Making an Attachment Device, the patented device is intended to perform the stretching of the strings of the molded devices of Bone 3,103,666. The patent recites that "[p]referably a plurality of the devices are molded together, the devices being connected by a rod . . . integrally joined to the cross-bars by necks. . . . [P]referably the devices are left interconnected in groups until they are used and are then severed successively as needed."
The following seven patents complete the list of those cited to the examiner by Dennison in its application and statement in support of special status, resulting in the accelerated examination referred to above: Kohnle 1,416,665, Kohnle 2,112,627, Flood 2,915,177, Marshall 2,937,464, Tamaccio 2,950,555, Marshall et al. 2,987,840 and Rieger et al. 3,273,705. The two Kohnle patents, Marshall 2,937,464 and Tamaccio require little discussion. They disclose paper tags manufactured in strip form, each tag to be separated from the strip in a machine. Tamaccio is perhaps the most relevant of these four patents in that the material joining each tag is slightly weakened by crenelation or perforation to allow easy separation. Marshall et al. 2,987,840 also discloses tags in strip form, but joins each tag to its neighbors by thin portions of the tag strip material at the top and bottom of each tag. Flood 2,915,177 is a continuation-in-part of Flood 2,825,162, cited by the examiner, and adds nothing of substance. Finally, Rieger et al. 3,273,705, Fastener, patented September 20, 1966, discloses an integrally molded plastic fastener with a needle at one end and an eyelet at the other. It is intended to be used either manually or with a gun. The attachments are assembled in clip form, with thin connections between the needles that are easily severed as the needle is projected out of the gun. While it is certainly more relevant as prior art than the six patents discussed immediately before it, it is not as relevant as Bone 3,103,666, cited by the examiner. Thus, of these twelve patents (the eleven cited by plaintiff in its application for accelerated examination plus Schumm) we need consider only those cited by the examiner in allowing the claims of the patent in suit, for the remaining references are merely cumulative with the Bone, Schumm and Flood patents cited.
The prior art references cited by defendant which neither the Patent Office nor plaintiff cited are: Spinner 2,435,561, Paxton 3,164,250, Spinney 3,285,404, Powers 3,348,669, Hauck 3,492,907, Keck 3,756,391, Rieger 3,185,367 and Japanese Utility Model Publication 45/XXXX-XXXXX ("JUMP"). Spinner, Paxton, Spinney, Powers and Hauck are cited by defendant in support of its statement that "[t]he prior art is replete with examples of the connection of a series of devices in a parallel array by connecting the devices at both ends thereof." Defendant's Proposed Finding No. 68. These five patents cumulatively disclose integrally-molded structures of plastic material having more than one thin, easily severable connection between the individual units in the assembly. In each of the patents, *412 both connections are of the same strength and are severed simultaneously. Clearly, none of these patents adds anything to Schumm's plastic fasteners, which are "connected by a plurality of narrow connecting portions integral with the enlarged end portions, which connecting portions may be easily severed to provide thus for a plurality of separate strips." Rieger 3,185,367 teaches a fastener gun which is intended for use in conjunction with the barbed needle/eyelet type of device disclosed in Rieger et al. 3,273,705, cited to the examiner by plaintiff. Thus, it adds nothing to our inquiry.
This brings us to the linchpins of defendant's argument that the claims of the patent in suit are invalid because they are obvious: Keck and JUMP. Keck 3,756,391, Fastener Assemblage, although patented on September 4, 1973, after the date of Lankton's alleged discovery, was filed on November 26, 1971. Fasteners, such as nails, are encircled by two (or more if the fasteners' length so requires) sleeves and held in a parallel, spaced relationship by thin webs connecting the sleeves. The web strips are designed to be one weaker than the other, to provide controlled fracturing. Because of the angular placement of a rib of plastic across the top web connection, the top connection is more prone to fracture when the nail is driven than is the bottom connection, which is traversed by a rib of plastic parallel to the top and bottom edges of the connecting strip. The top rib breaks by compression prior to the breaking of the bottom rib, which breaks by exertion of tension.
JUMP 45/XXXX-XXXXX is entitled A Pin Assembly for Attaching a Price Tag to a Cloth Product, and was published on November 20, 1970. The publication discloses attaching devices comprising a sharp, piercing bar, a string and an engaging piece, a group of which is held in assembly by adhesive applied to the outer edges of the bar and the engaging piece. The attachment of a tag to a piece of fabric is "executed" by use of an attaching device which JUMP fails to illustrate or describe. The attaching device is loaded with a clip of the pins, the sharp piercing portion of the pin is passed through the hole of a price tag, and "[a]t this moment" the leading attachment is separated from the pin assembly. The adhesive method disclosed in JUMP is claimed to be cheaper to manufacture than pin units connected "at one edge".
In addition to these seven patents and JUMP, defendant cites as prior art the entire plastic molding art, where, according to Samo's testimony, connections of the sort Lankton devised were commonly known. This aspect of the plastic molding technology, known as "gating", adds nothing specific to the content of the prior art, but certainly is part of the storehouse of skills and knowledge upon which one skilled in the pertinent arts would reasonably be expected to draw in devising a solution to a problem in the art. Having thus determined the scope and content of the prior art, our task now is to ascertain the differences between that art and the claims at issue.
*413 Differences Between the Prior Art and the Claims in Suit
It is undisputed that each of the elements of the claimed invention is disclosed in the prior art plaintiff does not claim that any single feature of the patented structure rises to the level of invention. What is claimed to be unobvious is the way each of these known elements has been combined to produce a useful and unexpected result. The combination may be unobvious even though each of its components is known in the prior art. Maclaren v. B-I-W Group Inc., supra, 535 F.2d at 1377-78. It is also undisputed that none of the prior art references, standing alone, discloses all of the features of Lankton's invention. Our task, therefore, is to examine the prior art and determine "whether the difference between the prior art and the subject matter in question `is a difference sufficient to render the claimed subject matter unobvious to one skilled in the applicable art . . .'" Dann v. Johnston, supra, 425 U.S. at 228, 96 S. Ct. at 1398.
The disclosures of Bone 3,103,666 meet the language of the first nine lines of Claim 1, which describes a clip of attachments in which the penetrating parts are connected by relatively sturdy frangible means which are adapted to be severed in an attaching device. However, the remainder of Claim 1, the dependent article claims and the method claims differ from Bone in that they describe and claim a clip of attachments whose paddle ends are joined by a connection described in the article claims and the method of manipulating the clip whose paddle ends are connected. The predecessors of the Bone patent, Flood 2,825,162 and Flood 2,915,177, disclose barstring-button attachments mounted in spaced relationship on an adhesive strip to provide for convenient and ready dispensing, adding nothing to Bone.
Schumm teaches plastic flexible ties joined in a strip by connections between the enlarged end portions of each tie. However, Schumm's connections, like those of the Spinner, Paxton, Spinney, Power and Hauck patents cited by defendant, and unlike those of the patent in suit, are not one more readily severable than the other, are not broken at different times, and are not broken by different means and forces in different operations. The connections in Schumm do not differ from one another in any of these respects.
The differences between the claims in suit and JUMP are in part similar to those between the claims and Schumm. JUMP, like Schumm, does not disclose connections of different strengths or connections that are to be severed sequentially or by different means. Thus, JUMP fails to meet the disclosures of the claims of the patent in suit relating to comparative severability, sequential severing and the method of severing each connection. JUMP is further different from Lankton's invention in that it discloses attachments which are individually molded and then secured together in a clip by adhesive on the edges of the penetrating parts and paddles; the attachments are immediately adjacent to one another, with no space between them.
In Keck, the connections are said to be of different strengths, but both are relatively strong, and both are broken in and by a machine, substantially simultaneously and in the same operation. As Crews acknowledged, unlike the patent in suit, Keck breaks the weaker connection before the stronger, Tr. at 3047, and requires almost simultaneous breaking of the connections in order to function properly. Tr. at 3051. Further, as plaintiff's expert, Rabins, testified, the connections must be relatively close together and be of closely similar strengths for Keck to operate. Tr. at 3553-54. Thus, Keck's counterpart to the patent *414 in suit's "securing means" (the second connection) is not remote from the first connection and possesses none of the other limitations found in the dependent article claims. The limitations of the method claims relating to sequential severing and the means by which each is severed are not met by Keck, and the same may be said of the rotation step contained in method Claims 15 and 17.
Thus, the essential differences between the claims in issue and the cumulative teachings of the prior art consisting of Schumm, Keck and JUMP, all of which utilize at least two connections between the individual attachments or fasteners, may be summarized as follows. First, the article claims of the patent in suit specify connections of substantially different strengths, while only Keck claims connections of different strengths. Keck's connections, however, need be of closely similar strengths in order for the device to function as claimed. Second, the patent in suit claims sequential severing of the connections, at different points in the attaching operation. Only Keck teaches sequential severing; but even in Keck, the connections are severed substantially simultaneously, as part of the same attaching operation. Third, the article claims of the patent in suit specify that one connection, the stronger of the two, is broken in and by an attaching device, while the second, weaker connection is broken externally of that device. None of the prior art references meets this limitation. Finally, Lankton's invention breaks each connection by applying different forces by different methods. The stronger connection is broken first, by the application of shear forces in the attaching device, and the second, weaker connection is broken by the joint application of torsion and tension, externally of the device. Again, only Keck breaks the connections by different forces. Keck first breaks the weaker connection by compression, then the stronger by tension. However, both connections in Keck are broken in the attaching mechanism.
The question is whether these differences between the prior art and the claims in suit are sufficient to render the invention disclosed by Lankton unobvious, or whether the differences are such that they would have been obvious to one skilled in the art. We define a person of ordinary skill in the applicable art as one with a scientific or engineering background possessing a practical knowledge of plastic molding and the attaching or fastening field. The ordinarily skilled person would be expected to know and understand the limits and capabilities of plastic molding, to be familiar with the properties of plastic materials, and to appreciate the processes involved in a marking operation using molded plastic attachments. Because of the breadth of the claims, we assume that the skilled craftsman also has a broad knowledge of various devices in use in the attaching or fastening field.
Defendant begins its arguments by citing examples in the tag attaching art which teach connecting as a means of avoiding tangling (Bone 3,103,666, the Flood patents, and the Merser applications, all of which are single connection devices); examples in the plastic molding art teaching a plurality of connections, placed at opposing ends of the devices (Spinner, Paxton, Spinney, Powers and Hauck, none of which specify the purpose of the connections, which appears to be to facilitate handling); the plastic molding art for its general teaching of strength and location of gates for particular purposes; Keck for teaching strength and sequence of severing of connections for particular purposes; and JUMP for its teaching of tag attachments joined at both ends by adhesive in order to prevent tangling. Relying primarily on the testimony of its expert, Samo, defendant argues that one skilled in the art, having before him these prior art references and faced with the problem of misalignment of the paddle ends of the Bone attachments, would connect the *415 paddle ends together by means of an integrally-molded plastic filament because that method would be less costly than if a separate operation, such as placing adhesive between the paddles, were performed. Further, Samo testified, the strength of the filament would be chosen to fit the need, using existing gating technology, and the connection would be placed where it would do the most good; that is, as close as possible to the paddle ends of the attachments, since the object is to prevent tangling of those ends. Finally, the hypothetical skilled artisan would know that the time at which the connection is broken or severed is dependent upon the other operations in separating an attachment from the clip, and that, if an object of the second connection is to prevent tangling, the connection should be broken as late in the attaching operation as possible. Tr. at 2363-78. Plaintiff characterizes Samo's testimony as "glib." While we would not go that far, we do think that his hindsight conclusions are not persuasive.
We find that Lankton's invention was not obvious to one of ordinary skill in the art. While all of the elements of the device were known in the prior art, Lankton combined them in a way that rose to the level of invention. JUMP, relied on heavily by defendant, is an unworkable patent that adds little, if anything, to the state of the art. It does teach a connection at the paddle ends of the attachments to avoid tangling, but the solution proposed by JUMP could easily lead one skilled in the art to abandon the idea of connecting the paddles because of its failure to solve the problem of providing a connection that is strong enough to hold the attachments together in handling but weak enough to allow easy breakage in use. JUMP simply discloses connections of the same strength at both ends of the attachments, fails to differentiate the methods by which they are broken or the stages in the attaching operation at which they are broken, and appears not to be adapted to use in an attaching device because of the lack of space between the attachments in the clip. It is vague and indefinite as to its exact teachings. These factors, in conjunction with the unworkability of the disclosed invention, might well have caused one skilled in the art who was considering the idea of connecting the paddles to avoid tangling to discard the notion and look for other solutions. Thus, rather than leading us to the conclusion that the claims in suit are obvious in the light of the prior art, JUMP may even be said to evidence the unobviousness of Lankton's invention in that it represents an unworkable and unsuccessful attempt at solving a recognized problem. (See discussion of secondary considerations, infra.)
Keck is vastly different from the claims in suit. Although it is considered part of the prior art because of the breadth of Lankton's claims, its teachings concern fasteners which bear little relation to the plastic flexible fasteners in issue here. What Keck teaches about connecting rigid nails is not readily transferable to plastic fasteners, and its teachings would provide scant guidance to one attempting to solve the problem of tangling of flexible filamentary devices. Even if Keck were more like the kind of fastener which is susceptible to tangling, the differences between it and the claims in suit, which we discussed above, are significant, and Keck's teachings consequently do not render the claims in suit obvious.
Utilizing hindsight, it might easily be said that Lankton's invention, which merely combined known, existing elements, was obvious. In addition to the prior art references discussed above, there is the plastic molding art, which teaches the selection of location and strength of connections as a matter of their function. It is admitted that connecting the opposite ends of the plastic attachments was an obvious expedient *416 to avoid tangling, Tr. at 11, and, defendant argues, once that decision had been made, the nature of the connection was a matter within the ordinary skill of one practicing the gating art. This argument fails to appreciate the problem Lankton solved and the extent of the insight required to solve it. A paddle connection would be useless unless it were weak enough to be broken easily in use yet strong enough to withstand ordinary dislocative forces. The key to the solution lay in envisioning what would take place when the leading attachment in a clip was associated with an object and then withdrawn if there were a connection between its and the next adjacent paddle end the leading attachment would rotate relative to the clip, thereby exerting torsional forces on the connection, and as the gun was pulled away, the connection would also be subjected to tension. Perceiving these relationships required inventive genius and a connection that took advantage of that perception was the embodiment of that inventiveness. Not the plastic molding art and not the cumulation of the prior art references discussed above made the invention obvious.
B. Secondary Considerations
Plaintiff contends that the commercial success of the product, the long felt need in the industry for such an invention, the failure of others to arrive at a solution, and the recognition afforded the invention by the industry provide further proof of its unobviousness. Although such "secondary considerations" without invention will not tip the balance of patentability, Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., supra, 340 U.S. at 153, 71 S. Ct. 127, and will not suffice to demonstrate lack of obviousness where the differences between the prior art and the claims in suit are insubstantial, where the obviousness of the invention based on the Deere objective criteria is unclear, as it is here and as it "virtually always" is, Lerner v. Child Guidance Products, Inc., supra, 406 F. Supp. at 565, secondary considerations provide more reliable signposts of patentability than do a priori evaluations aided by hindsight, id., 406 F. Supp. at 566, and are relevant to the determination of obviousness. Dann v. Johnston, 425 U.S. 219, 230 n.4, 96 S. Ct. 1393, 47 L. Ed. 2d 692 (1976); Graham v. John Deere Co., supra, 383 U.S. at 17-18, 86 S. Ct. 684; Formal Fashions, Inc. v. Braiman Bows, Inc., 369 F.2d 536, 539 (2d Cir. 1966).
Defendant argues that the commercial success of the "Super Swiftachments" has not been proven to be the result of the superiority of the product, but rather that its success was attributable to Dennison's dominant position in the market for attachments of the old type (with unconnected paddles), an aggressive advertising campaign to sell the connected paddle devices, and Dennison's substitution of the newer device for the older. Defendant claims that it had no choice but to follow plaintiff's lead in marketing the new devices. Defendant contends that there is still a substantial market for the unconnected devices, as evidenced by the testimony of Francis G. Merser, a Dennison marketing director, that the unconnected devices are used to meet price competition. From the fact that customers are willing and ready to accept the unconnected attachments defendant concludes that the success of the Super Swiftachment device results solely from their substitution for the unconnected devices. This conclusion overlooks the fact that the Super Swiftachments, although cheaper to manufacture, sell at a higher price than the Swiftachments but are nevertheless generally more in demand than the latter. Only by lowering the price of the Swiftachments still further below that of the Super Swiftachments are customers enticed by the economics of the situation to choose the less desirable unconnected devices. Whatever market exists for the *417 unconnected attachments is exploited only by sufficiently lowering the price.
Moreover, there was ample proof that the success of Lankton's invention was directly related to its contribution to the state of the art in the marking industry. Plaintiff produced letters from defendant's files of two kinds, both of which indicate the demand for connected paddles: inquiries from customers requesting connected paddles, including returns of unconnected shipments to be exchanged for connected, and letters from defendant to its suppliers requesting connected paddle devices. Plaintiff's Exhibits 44 and 44A. Alan Clements' attempts to explain away the significance of these letters, in terms of the necessity of following the leader, were unconvincing. In addition, both plaintiff's and defendant's sales statistics show a substantial demand for the connected paddle devices. All of defendant's # 5, # 22 and # 35 (fine) devices have connected paddles, as does 80% of its # 3 (regular) devices. Tr. at 2047-54. Alan Clements testified that "some" distributors want only connected paddle devices. Tr. at 2054. These facts indicate a strong preference on the part of customers for connected paddle devices, which cannot credibly be said to be the creation of plaintiff's advertising. They also indicate the commercial success of the new device in that Clements, by its own estimation Dennison's main competition in the field, Tr. at 2080, was compelled by its customers to furnish the superior device. Plaintiff, too, was forced to supply a connected version of its fine fastener by a major customer who threatened to buy Clements' connected fine fastener unless Dennison would manufacture one. Tr. at 342, 544-46, 690. An industry witness testified that he much preferred the connected devices and would not return to the unconnected devices under any circumstances. Tr. at 1670. The conclusion that the success of the Super Swiftachment was due to its superiority over the prior devices and was not solely the result of plaintiff's market position and strategy is thus amply supported by the proof.
Long-Felt Need and Failure of Others
While the Swiftachment devices introduced in 1965 represented a significant advance over the then-existing state of the art, plaintiff contends that these devices were plagued by a number of problems which curtailed their efficiency and usefulness. First, the strings or elongated portions of the attachments tended to tangle with strings in the same clip or in other clips when they were packed in a box, requiring the operator to pause to untangle the strings before proceeding with the marking operation. A clip had to be freed from other clips before it could be inserted in the attaching gun and tangling of strings within the same clip would cause the T-bars to depart from a parallel orientation, causing the gun to jam. Second, plaintiff argues that with the advent of out-of-the-mold stretching in 1966 came the problem of unstretched strings, which occurred when the stretching apparatus would fail to grasp a paddle that had for some reason gone astray from the parallel orientation in which it was molded. The presence of an unstretched string on a clip also cost the operator time to remove the clip from the gun and, since clips with unstretched strings were unacceptable to customers and were discarded by Nylon Products, Dennison's supplier (see n.1), unstretched strings represented a production loss. Third, rotation of a paddle from its proper parallel orientation caused the T-bar end to become misaligned, which, if substantial, caused the guns to jam, again requiring the operator to pause and correct the condition. It is urged by plaintiff that the existence of these problems and the consequent unsuccessful efforts by others to solve them further evidence the unobviousness of Lankton's invention, which was successful in solving them. Defendant responds that the problems were greatly exaggerated by plaintiff *418 or that they were solved by other methods or improvements.
Defendant's contentions with respect to the tangling problem need not detain us long. The existence of the tangling problem was amply demonstrated at trial by the testimony of Dennison and Nylon Products personnel which established that customer complaints of tangling had been received with regularity from the introduction of the Swiftachment devices. Tr. at 267-68, 904, 922. See also Ex. 96. Alan Clements testified that he had discussed the tangling problem with his suppliers. Tr. 2101-02. Moreover, the deposition testimony of witnesses involved in the marking field for major retailers documents the existence and severity of the tangling problem. The fact that few records of customer complaints were produced is of little consequence, since Merser testified that such nonessential records were disposed of after a relatively short period of time, Tr. at 267, and Lankton testified that most of the complaints he received from Dennison were verbal. Tr. at 905. Both witnesses were fully credible. The fact that no customers were lost from 1965 to 1972 due to the tangling problem is also unavailing to defendant because there were no superior alternatives to the Swiftachment type of fastener during that period.
Defendant next argues that even if tangling was a problem, it was viewed as a packaging problem and that, therefore, it provided no incentive to devise a structural solution as Lankton did. The evidence fully supports the conclusion that plaintiff, defendant and their competitors perceived tangling as a problem and attempted to solve it by packaging techniques such as placing tissue paper between clips of attachments in their boxes, rotating the clips 90° or 180° relative to one another, or placing them in separated packages, eggcarton style. Tr. at 130-36, 480, 904-06, 1011-12, 1851-56, 2090-91, 2506-08. See also, e. g., Exhibits 51, 53, 54. However, that tangling was viewed as a packaging not a structural problem further supports the conclusion that Lankton's solution was ingenious and unobvious he possessed the vision to consider alternatives that others had ignored.
The unstretched string and jamming problems present somewhat different situations. Unstretched strings result from a string twisting or crossing over other strings, which in turn is caused by the clip sticking in the mold or being subjected to impact during handling. Both the sticking and the handling difficulties can be and were successfully remedied by improvements in molding and handling techniques. Similarly, jamming of guns can be and was alleviated by improvements in gun design. The need for a solution to the unstretched string and jamming problems, therefore, was not acute, since those problems had been successfully solved by other methods. In contrast, packaging techniques fell short of solving the tangling problem, because those techniques were labor intensive and, hence, too expensive to be employed by a domestic manufacturer. Tr. at 904-06. Moreover, the structural solution is a more economical solution to the tangling problem, wherever the manufacturer is located, since no additional operation, such as careful packaging by hand, is required. Tr. at 2365-66. Thus, unlike the tangling problem, the problems of unstretched strings and jamming provide little support for the unobviousness of Lankton's invention.
As further relevant evidence of unobviousness and thus validity, plaintiff points out that of all of its competitors in the plastic fastener marketplace, only defendant fails to respect the patent in suit. In appropriate circumstances, public acquiescence in the validity of a patent may be evidence of that validity, particularly where, as here, the patented device enjoys a great deal of commercial success. Georgia-Pacific Corp. v. United States Plywood Corp., supra, 258 F.2d at 133-34; see Note, Subtests of "Nonobviousness": A Nontechnical Approach to Patent Validity, 112 U.Pa.L.Rev. 1169, 1178-80 (1964). Although the economic incentive to copy plaintiff's device certainly existed, only defendant *419 has persisted in marketing a device that allegedly infringes plaintiff's patent. Other manufacturers and importers have sold connected paddle devices in the past, but discontinued doing so upon being notified by plaintiff of the existence of the patent in suit and of plaintiff's belief that its patent was being infringed. Exhibits 48, 49 and 118. It is significant that competitors who were in better economic positions than defendant to challenge the patent in suit have not marketed connected paddle devices. Two competitors were divisions of large companies, easily able to finance a patent litigation, and a third was, in Alan Clements' estimation, better able than Clements to fight Dennison, Tr. at 2042; yet none of these competitors has marketed a connected paddle device, despite the commercial success that these devices enjoyed and the customer demand that they engendered. Their and other competitors' acquiescence is thus evidence of the validity of plaintiff's patent.
In sum, we find that the secondary considerations of commercial success, long-felt need, failure of others and industry recognition lend support to our conclusion that plaintiff's patent is not obvious in light of the prior art when judged by the objective, primary criteria set forth in Deere, supra. As a counterbalancing secondary consideration, defendant argues that the contemporaneous, independent development of connected paddle devices by another manufacturer is evidence of the obviousness and invalidity of plaintiff's patent.
Contemporaneous Independent Development
The contemporaneous, independent development of an invention substantially similar to that of a patent in suit is evidence, although not conclusive evidence, of the obviousness of the claims in suit. Lerner v. Child Guidance Products, Inc., supra, 547 F.2d at 31; Reeves Brothers, Inc. v. U.S. Laminating Corp., supra, 417 F.2d at 872. The weight and relevancy of the evidence is dependent upon a number of factors, including the similarities between the two conceptions, the time between the identification of the need for a solution to a problem and the solution, and the sequence of and time between the patented conception and the alleged independent conception. Defendant claims that its Japanese supplier, Japan Bano'k Co., Ltd. ("Bano'k"), developed two connected paddle devices, the Hi-Lock and fine attachments, at approximately the same time as or before Lankton developed his device and that, therefore, these developments are significant evidence of the obviousness of Lankton's connections.
Alan Clements testified that he first saw five-inch attachments with holes in the paddles (Hi-Lock) joined at the paddle ends by partial cylindrical portions of plastic material extending from the circumferences of the holes in December, 1971, during a visit to Japan, but defendant offers nothing to corroborate that testimony. However, whether Bano'k demonstrated to Alan Clements connected-paddle Hi-Locks in December of 1971 is a question we need not decide in light of the evidence that such attachments existed four months later. By letter of April 28, 1972, Bano'k transmitted to Clements an unstretched sample of 5" Hi-Lock devices "of which heads are connected (sic) each other to make the handling easy." Exhibit AU. On June 22, 1972, Bano'k sent to Clements 5,000 Hi-Lock devices for display in a July, 1972 trade show in Los Angeles. Exhibit BA. Although the June transmittal letter does not state whether the paddles of these devices were joined, the brothers Clements all testified that the Hi-Locks that were displayed at the trade show had connected paddles. The only evidence *420 offered by plaintiff to contradict this testimony was a report on the July, 1972 trade show, prepared by a former sales representative of Dennison, transmitting Clements' price lists obtained at the show, Exhibit 136, and his testimony from this report. The report states that Clements was about to market Swiftachment-type attachments with round runner bars and asks the Dennison home office about another device displayed at the Clements booth. The failure of the report to mention the connected Hi-Locks, plaintiff contends, is evidence that this device was not displayed at or in existence at the time of the July, 1972 show. We find this report and the testimony from it unpersuasive evidence that defendant did not display connected Hi-Locks at the show, and it is more than outweighed by the letters referred to above and the testimony of the Clements brothers. However, while we think that there is ample evidence that defendant showed as early as July, 1972 a connected-paddle Hi-Lock, we conclude for the reasons that follow that the Hi-Lock device was not such an independent contemporaneous development as to affect our conclusion that Lankton's invention was unobvious.
Defendant admits that the paddles of the Hi-Lock device are joined by "flash", which is excess plastic in a mold, generally undesirable and generally indicative of an out-of-control molding process. Tr. at 916-19. Although defendant takes issue with the undesirability of flash and argues for the existence of a "flash-gate art", Tr. at 2347-48, we are unconvinced that flash is reliable as a means to connect and align the paddles. Tr. at 941-42. While we think that plaintiff overstates the problems associated with the use of flash to join the paddles of the Hi-Lock devices (we refer to its claims that the flash is too strong to be useful and that, when broken, it leaves vestiges on the paddle faces that might tear garments), the weight of the evidence is that the flash connection was not a feature intentionally developed by Bano'k. First, in addition to the above general observations about the undesirability and lack of control of flash, we note that, except for the letter of April 28, 1972, transmitting unstretched samples, Bano'k makes no mention of the connection between the paddles of the Hi-Lock even that letter refers to the connection only as an aid in handling the devices. Bano'k's advertisements nowhere refer to the connected paddle feature. Exhibits BF-2a, BF-2b. Defendant argues that the photographs in Bano'k's September, 1972 advertisements show the paddles connected, but we do not find those photographs conclusive on this issue. See Exhibits BF-1 through 7. Second, the connection appears to be more of a detriment than an advantage. In order to use the attachment in the only manner in which it is depicted in Bano'k's ads, that is, in the form of a loop, with the T-bar end passed through the hole in the paddle, it is necessary to detach the paddle end from the clip before the leading attachment is used. Tr. at 2219-20, 2540-41. Other uses of the Hi-Lock, demonstrated by defendant at trial, were seemingly not envisioned by Bano'k and were somewhat contrived. Tr. at 3235-40, 3280. Third, we note that its 5" attachments without holes in the paddles were not connected. Therefore, it is a fair and compelling conclusion that the flash connection was a molding accident, a mistake, upon which Bano'k attempted to capitalize in its letter of April 28, 1972 by referring to it as a handling aid. Even if the molding imperfection persisted beyond the July, 1972 trade show, and we have no convincing evidence on either side of this issue, it would not provide a basis for concluding *421 that Lankton's conception was obvious, for there still is no evidence that Bano'k intentionally developed or looked upon the connection as significant in any way, or, indeed, as anything but an inconvenience, as witnessed by the absence of its mention in Bano'k's promotional material. In order to rise to the level of an independent contemporaneous conception that is evidence of obviousness, the independent conception must be something that is intentionally developed, not happened upon by accident. Defendant has, therefore, failed to persuade us that the Hi-Lock provides any evidence of the obviousness of Lankton's invention.
Fine Fabrics Fasteners
Defendant also claims that Bano'k developed a connected paddle fine fabric fastener contemporaneously with, perhaps even before, Dennison's development of its connected paddle device.
Although Lankton is conceded to have conceived of his paddle connections in February of 1972, the devices were subjected to in-house testing before being sent to Dennison's customers for test marketing. The evidence shows that in early September, 1972, Dennison sent 10,000 five-inch attachments with connected paddles to each of five locations, two in New York City and three in Massachusetts. Exhibit 76. Towards the middle of September, test marketing began on a connected paddle version of Dennison's three-inch attachments, which were tested in significantly larger quantities and at many more test sites than the five-inch. Exhibit 74. The tests were to be conducted on a "confidential disclosure basis", Exhibit 76, but the attachments were to be used in the course of normal marking operations. Tr. at 310. Because there was no attempt (nor could there be, as a practical matter) to keep the new attachments secret during the testing process, plaintiff argues that defendant or one of its distributors could very well have seen the attachments in use in a store, in the marking room or attached to a garment on display. One who was familiar with the marking field could see from an attachment holding a tag to a garment that it had been joined at the paddle to its adjacent attachments and, from the dimensions of the residual protrusions and knowledge of normal paddle spacing, would be able to fashion a connection performing in the same manner as Lankton's. An existing mold could be modified within a week's time to produce a sample of a connected paddle clip of attachments. Tr. at 1001-02. On September 26, 1972, approximately three weeks after Dennison began test marketing its five-inch connected paddle attachments, Bano'k transmitted to Clements "samples of newly developed fine size nylon pin, 7mm, of which feature is the slightly joined head in order to avoid the entanglement in the actual use." Exhibit BG.
While this letter is the first documented reference to Bano'k's fine fabric connected paddle fastener, Alan Clements testified at trial that he thought, but was "just not sure," that he had seen such fasteners in Japan in early December, 1971, Tr. at 2182-83, and at his deposition that he "could have possibly seen the pins with the joined or barbs with the joined paddle," Tr. at 2184, at that time. He also testified that he saw the connected paddle fine fasteners again at a trade show in Holland in March, 1972, and a third time in Los Angeles in July, 1972, when a Bano'k representative brought some samples to him to solicit orders. Because the first documented appearance of the Bano'k connected paddle fastener occurred after Dennison had begun its test marketing and so exposed its new device to the public, albeit in limited quantities, the question of the independence of Bano'k's development warrants particularly close scrutiny. The only corroboration defendant offers for Alan Clements' claim that he may have seen connected paddle fine fasteners in Japan in December, 1971, consists of a price list sent by Bano'k to Clements on December 6, 1971, Exhibit AL, and a letter from Bano'k confirming shipment of an order, *422 dated December 16th of the same year. Exhibit AM. Both exhibits refer to fine attachments of the 6mm and 15mm sizes, but neither mentions connections between the paddles as a feature of the fasteners. A similar lack of corroboration exists for both the March and July, 1972 encounters at which Alan Clements claims he was shown connected paddle fine fasteners by Bano'k. We do not doubt that Alan Clements met with Bano'k representatives in Holland in March, 1972 (see Exhibits AP, AQ and AS), and in Los Angeles in July, 1972 (see Tr. at 1952), but there is no evidence other than his own testimony that he was shown connected paddle fine fabric fasteners on these occasions. Alan Clements' obvious self-interest in establishing a date for Bano'k's development of connected paddles prior to Dennison's test marketing also creates doubt in our minds as to the credibility of his totally uncorroborated testimony at trial.
On the other hand, the evidence clearly supports an inference that Bano'k did not develop its connected paddles until September of 1972. First, there is the lack of correspondence between Bano'k and Clements regarding the connected paddle feature. Defendant was Bano'k's sole North American distributor in 1972, yet was not advised in writing of the connected paddle development until late September, 1972. Given the pattern and regularity of correspondence between the two firms, had Bano'k developed the feature earlier than September, 1972, as defendant claims, it is highly likely that Bano'k would have referred to the development in writing, touting it and soliciting orders for the new product. The absence of such a writing is conspicuous. Second, in Bano'k's advertisements in several Japanese trade journals published on September 4, 1972, at the time when Dennison's test marketing was just getting underway, there is no mention of a connection between the paddles of the fine fasteners. Exhibit BF. Yet only three weeks later, Bano'k sent to Clements "samples" of "newly developed" connected paddle fine fasteners, pointing out that advantageous feature. Exhibit BG. We find defendant's argument that the advertisements did not specify connected paddles because that feature was not favored in the Japanese domestic market unpersuasive. Third, that defendant did not order any of Bano'k's fine fasteners with connected paddles until December, 1972, a full year after Alan Clements claims to have seen the attachments in Japan, casts further doubt on defendant's assertion that Bano'k had developed connected paddles as early as December, 1971. Rather, the order came in relatively close proximity to Bano'k's transmittal of samples in late September, 1972. Clements did not show the connected paddle fasteners until February, 1973, at the same show at which plaintiff's connected attachments were unveiled. The question that defendant has failed to answer satisfactorily is why, if samples had been available since December, 1971, did it wait until a year later to order and until February of 1973 to show the connected paddle fasteners? In a highly competitive industry such as this where new developments are quickly exploited, the most plausible explanation is because Bano'k had not developed connected attachments until late in September, 1972, after Dennison had commenced its test marketing.
In an effort to bolster his claim that the fine fasteners he saw in Japan in December, 1971, had connected paddles, Alan Clements testified that his company had never sold unconnected fine attachments, Tr. at 1887, nor, to the best of his knowledge, had Bano'k ever manufactured any. Tr. at 1900-01. This testimony was contrary to his deposition testimony concerning the tangling problem associated with fine fasteners, Tr. at 2093-2100, and also contrary to Stanley Clements' deposition testimony on the same topic. Tr. at 816. We refer again to the price lists furnished by Bano'k in December, 1971 and July, 1972, both of which list fine barbs but neither of which mentions connected paddles. Both lists predate Bano'k's first specific reference to connected paddles in September, 1972. Thus, it is clear that Bano'k manufactured unconnected fine fasteners, and it is most probable *423 that the fine fasteners Alan Clements saw in Japan in December, 1971 were unconnected. See Exhibit 127.
As we have noted, the evidence that defendant appropriated plaintiff's invention is inferential, consisting primarily of defendant's unconvincing attempts to prove that Bano'k had developed connected paddles prior to the initial documentation of September 26, 1972. Further evidence that Bano'k's development was not independent of Dennison's is that Clements did not notify Bano'k that plaintiff had shown connected paddles at the February, 1973 trade show. This lack of correspondence is telling, since Clements had been fastidious in keeping Bano'k informed of its competitors' latest developments. Even if Bano'k had independently developed a connected paddle, the fact that Dennison, the acknowledged leader in the attaching field, now had the same feature would certainly have been of interest to Bano'k unless Bano'k already knew of Dennison's development, as had Clements, since shortly after Dennison's test marketing had begun. Defendant offered no direct evidence, other than the testimony of its own interested witnesses, to refute the inference of appropriation, even though such evidence was available in the form of the testimony of industry witnesses to whom the connected fine fasteners were allegedly shown in July, 1972. We, therefore, conclude that defendant has not met its burden of proving that Bano'k's development was independent of Dennison's.
However, even if we were to find that Bano'k's development was independent of Dennison's, the weight of that development as secondary evidence of obviousness would not be substantial, due to the time that had passed between the identification of the need for a solution to the tangling problem and the solutions. Cf. Lerner v. Child Guidance Products, Inc., supra, 547 F.2d at 31. The Swiftachment-type fastener and its tangling problem had been in existence since 1965 and the solutions were not devised until seven years later. Under these circumstances, it is difficult to say that the solutions were obvious.
The net result of these secondary considerations, thus, has not been to cast doubt upon our conclusion that the claims in suit are not obvious when judged by the criteria established in Deere, supra, rather, they have reinforced that conclusion. On the basis of those primary criteria and secondary considerations, we conclude that defendant had failed to meet its burden of proving that the claims in suit are obvious.
VII. Deceptive Intention
35 U.S.C. § 288 reads in pertinent part:
Whenever, without deceptive intention, a claim of a patent is invalid, an action may be maintained for the infringement of a claim of the patent which may be valid.
Defendant argues that, even if not all of the claims of the patent which are in issue in this suit are held to be invalid for the reasons discussed above, this suit may not be maintained on the valid claims because the invalid claims were obtained with deceptive intention on the part of Lankton. Specifically, defendant cites Lankton's deceptive intention in prosecuting those claims which were broadened beyond the scope of his conception and in claiming the embodiment of Claim 13, which substitutes an adhesive for Lankton's conception of a filamentary connection. We have already dealt with the first part of defendant's argument in our discussion of defendant's contention that the claims of the patent in suit are invalid for "overclaiming". We found the argument without merit in that context and we so find it here. As to defendant's argument based on the presence of Claim 13 in Lankton's patent, we note first that defendant has not been accused of infringing this claim. Its place in this lawsuit is limited to its role in defendant's argument that its invalidity taints the *424 entire patent. Thus, we will decide whether it is valid or invalid solely on the grounds put forth in connection with defendant's argument of deceptive intention. We do not decide its validity measured against any other argument or criteria.
Defendant's argument is that, because Lankton did not "invent" the adhesive embodiment contained in Claim 13, it is therefore an invalid claim; and since Lankton obtained the patent in suit knowing that it contained an invalid claim, suit may not be maintained for infringement of its otherwise valid claims. It is uncontested that the specific idea of putting adhesive between the paddles to perform the same function as the filamentary connection was not Lankton's. Rather than claim only Lankton's conception, Dennison assigned one of its engineers, Arnold Bone, to examine Lankton's invention and to develop, in collaboration with Lankton, additional ways of performing the function of Lankton's filament. It was probably Bone who devised the adhesive embodiment, building on Lankton's invention and having the benefit of its teachings, the resulting connection possessing the same functional characteristics as Lankton's filament. We conclude that Bone's contribution, as found in Claim 13, was an obvious variant on Lankton's conception and was non-inventive. "To claim inventorship is to claim at least some role in the final conception of that which is sought to be patented." Mueller Brass Co. v. Reading Industries, Inc., 352 F. Supp. 1357, 1372 (E.D.Pa.1972), aff'd, 487 F.2d 1395 (3d Cir. 1973). What was patented was a construction or connection with the characteristics Lankton envisioned; that is, a connection that would be strong enough to withstand dislocative forces during handling, but would break easily during and as part of a normal attaching operation as a result of the combination of torsion and tension that is visited upon the connection during the attaching process. In understanding that the paddles could be connected by connections having these characteristics, Lankton had no assistance the concept was his alone. Bone's role lay in taking Lankton's conception and devising alternative ways of implementing it, hardly requiring any inventive genius, since the adhesive Bone came up with had the characteristics deemed essential by Lankton. Bone's "improvement", if any, was not "so significant as to amount to `a complete invention' in and of itself." Id., 352 F. Supp. at 1373. Rather, the inventive genius resided in Lankton alone. See Technical Tape Corp. v. Minnesota Mining and Manufacturing Co., 143 F. Supp. 429, 437 (S.D.N.Y. 1956), aff'd, 247 F.2d 343 (2d Cir. 1957), cert. denied, 355 U.S. 952, 78 S. Ct. 537, 2 L. Ed. 2d 529 (1958).
Thus, because we do not think that the adhesive embodiment of Claim 13 amounted to a separate and distinct invention, but was within the parameters of Lankton's conception, Claim 13 should not be declared invalid on this basis. A fortiori, there was no deceptive intent on Lankton's part in obtaining this or any other claim. However, even if we were to find that Claim 13 represented an inventive step on Bone's part, we assume without deciding that he could be added to the patent in suit as a joint inventor. 35 U.S.C. § 256; Rival Manufacturing Co. v. Dazey Products Co., 358 F. Supp. 91, 100-02 (W.D.Mo.1973).
Having determined that the claims in suit are valid and that this action may be maintained under 35 U.S.C. § 288, we turn now to the question of infringement.
Defendant is accused of infringing claims relating to plaintiff's structure and to the method of manipulating it. As we previously noted, plaintiff, in the Pre-Trial Order, charged infringement of article Claims 1, 2, 3, 4, 5, 7, 9, 10 and 11 and method Claims 14, 15, 16 and 17; but in its Trial and Post-Trial Briefs plaintiff stated that it was confining its case to Claims 1, 2, *425 4, 5, 7, 10, 11, 16 and 17. Since plaintiff has offered no proof of the infringement of Claims 3, 9, 14 and 15, we make no finding as to whether they are infringed by defendant's devices. We consider first the article claims and then the method claims, which present somewhat different questions of law.
A. Article Claims
Claims 1, 2 and 11
Defendant concedes that these broad claims, if read literally, are infringed by its structures. It asserts, however, that these claims are broad enough to read on "every form" of paddle connection and are consequently too broad to be valid, an argument which was disposed of in Section V, B, of this opinion. But in the context of infringement, the argument merits further analysis.
The argument is that, in assessing whether these claims are infringed by defendant's structures, resort should be had to all of the teachings of the specification to determine exactly what structures would or would not infringe the claims. To further define the securing means claimed, plaintiff has relied on the strong-in-tension, weak-in-torsion property of its claimed structures, a property not found in Claims 1, 2 and 11, but in the specification. Defendant contends that plaintiff may not pick and choose from the specification those portions which suit its purpose, but must refer to all of the teachings and limitations contained in the specification. The teaching defendant has in mind is that of Col. 6, lines 18-36, which states that "[t]he normal spacing between attachments in a clip, usually about 0.04 inch, is far too large to produce a connecting means [which can be broken relatively readily when twisted]". (Emphasis added.) This limitation, it is argued, should apply to all of the claims in suit and because defendant's connections span the full distance between paddle faces, a distance "far too large", there is no infringement of the claims.
The specification is directed primarily toward the construction actually conceived by Lankton a "thin and short filamentary connection", Col. 3, line 35, a "special construction" designed to provide a "tension-resisting and torsion-sensitive connection" despite the normal distance between attachments in a clip which "militates against the attainment of such a . . characteristic." Col. 3, lines 46-54. However, Claims 1, 2 and 11 are broader than the specific embodiment favored by Lankton, which is claimed in article Claims 6, 8, 12, 18, 19, 20, 21 and 22, which defendant is not accused of infringing. Because the claims, not the specification, define the limits of the patent monopoly, Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., supra, 340 U.S. at 149, 71 S. Ct. 127; Maclaren v. B-I-W Group Inc., supra, 535 F.2d at 1372, it would be inappropriate for us to read into these claims limitations contained in the specification. Just as a patentee may not import into the claims limitations contained in the specification in order to preserve the validity of the claims, one accused of infringing a claim may not avoid infringement by arguing that a broad claim should be limited by the specification, particularly where, as in this case, there are narrow claims specifically drawn to those limitations. Thus, Claims 1, 2 and 11 may be infringed by a structure not meeting the so-called "negative teaching" of Col. 6, lines 18-36. On the other hand, the "strong/weak" characteristic relied upon by plaintiff is the essence of Lankton's invention, and serves to explain, not limit, the securing means described in all of the claims. That characteristic may be achieved by the short filament/protrusion structure of the narrower claims, or by any *426 other structure meeting the elements of Claims 1, 2 and 11. Because, as defendant admits, its structures meet each of the elements contained in these claims, we find Claims 1, 2 and 11 of the patent in suit infringed.
Claims 4 and 10
Defendant contends that its connections do not meet the elements of Claims 4 and 10, basing its argument on the meaning of the word, "twist". As used in these claims, defendant argues, twist means rotation alone; and, thus, defendant's connections, which require a small pull in addition to rotation to break them, do not infringe the claims. This argument is meritless. A dictionary meaning of twist includes "to pull off, turn, or break by means of a turning strain . . . to cause to move with any of various turning motions (as by pivoting, revolving, or spiraling) . . ." Webster's Third New International Dictionary 2473 (1964) (emphasis added). The definition is not confined to a purely rotating motion. But we do not rely solely on the dictionary meaning of the wordinstead, it is proper to look to the specification to ascertain what the inventor had in mind when he used the word, twist, in these claims. Canaan Products, Inc. v. Edward Don & Co., 388 F.2d 540, 544 (7th Cir. 1968); Kaz Mfg. Co. v. Northern Electric Co., supra, 412 F. Supp. at 477; see United States v. Adams, supra, 383 U.S. at 49, 86 S. Ct. 708. It is clear from Figures 3 and 4 and their description in the narrative (Col. 4) that Lankton differentiated between "rotate" and "twist" and that the latter included the former plus some degree of pull. Figure 3 depicts the leading attachment as "rotating with respect to the other attachments still remaining on the clip," while Figure 4 depicts "the leading attachment having been twisted off from the clip" with the remainder of the clip moving away with the attaching device. The act of moving away from the leading attachment provides the pull encompassed within the meaning of the word, twist. Indeed, the "continuous movement" (Col. 6, line 12, is one place where these words are used) of the operator during the marketing procedure provides the pull required to break the connection. Further support for this interpretation of twist is found in the testimony of Samo, Tr. at 2486, that breaking by rotation alone was not "practically" feasible. This lack of practical feasibility provides some indication of what Lankton, who possessed a great deal of expertise in the pertinent arts, meant when he used "twisted" instead of "rotated".
There is ample evidence, and defendant does not dispute, that its connections otherwise possess the characteristics described in Claims 4 and 10. Thus, these claims too are infringed.
*427 Claims 5 and 7
There is no dispute that defendant's structures infringe these dependent claims, nor is there any argument as to the meaning of these claims. Accordingly, Claims 5 and 7 are held to be infringed.
B. Method Claims
Defendant proffers two arguments that it has not infringed the method claims of plaintiff's patent. The first is that plaintiff has failed to prove that defendant has infringed under the applicable statutory provisions; the second is similar to arguments made with respect to the article claims that defendant's connections do not meet the language of the claims.
Some background is required to appreciate the statutory argument. We found defendant guilty of infringing the article claims of the patent by virtue of the fact that defendant sold the infringing devices in this country. 35 U.S.C. § 271(a). However, because defendant does not itself use the infringing devices, it cannot be said directly to infringe the method claims of the patent, which cover a "method of attaching" whereby the connection is broken in the course of and in accordance with the operation set forth in the claims. To be a direct infringer of the method claims, defendant must be found to have used the attachments in question in the manner prescribed in the method claims. Cf. B. B. Chemical Co. v. Ellis, 117 F.2d 829, 833-34 (1st Cir. 1941), aff'd, 314 U.S. 495, 62 S. Ct. 406, 86 L. Ed. 367 (1942). Therefore, the mere sale of the devices does not make defendant an infringer of the method claims. Id. If defendant has infringed the method claims, it has done so as a contributory infringer or by inducing infringement.
A "contributory infringer" is one who
sells a component of a patented machine, manufacture, combination or composition, or a material or apparatus for use in practicing a patented process, constituting a material part of the invention, knowing the same to be especially made or especially adapted for use in an infringement of such patent, and not a staple article or commodity of commerce suitable for substantial noninfringing use, ...
35 U.S.C. § 271(c). The clear intent of this subsection is to make liable for infringement one who supplies a component which is not patented to someone else who uses the component to infringe a patented machine or process. If the criteria of § 271(c) are met, the supplier of the component will be liable as a contributory infringer, a concept derived from that of a joint tort-feasor. 7 Deller, Walker on Patents, § 515 at 180 (2d Ed. 1972). Although the issue of contributory infringement will not ordinarily arise where, as here, the thing supplied is itself patented, Federico, Commentary on the New Patent Act 53, 35 U.S.C. preceding § 1 (1952), we think it appropriate to consider this issue in the context of these method claims which are not directly infringed by defendant.
Defendant's argument that it is not a contributory infringer is that the devices it furnishes to its distributors who in turn sell them to customers who use the devices are "suitable for substantial noninfringing use" and thus their sale does not give rise to defendant's liability as a contributory infringer under § 271(c). It appears (and defendant does not dispute) that defendant and its devices meet the other criteria of § 271(c). See Federico, Commentary on the New Patent Act, supra, at 53. As a basis for its argument that the accused devices are suitable for substantial noninfringing use, defendant cites instances of markers breaking the connections between the paddles before inserting the attachedpaddle *428 clip into the attaching gun. One of plaintiff's industry witnesses who was employed by a large retailing concern testified that, in the company of defendant's attorney, he saw a marker break the connections prior to using the clip, Tr. at 184-86, and Stanley Clements testified that he "occasionally" saw markers pre-break the connections. Tr. at 3275-77. Such occasional aberrant use of a product that is clearly designed to be used in a particular manner, namely, in the manner specified in the method claims, does not make defendant's device "a staple article or commodity of commerce suitable for substantial noninfringing use." 35 U.S.C. § 271(c). Alan Clements himself characterized the idea of breaking the connections before inserting the clip into the gun as "ridiculous", Tr. at 2548, and we agree with that characterization. We find, therefore, that defendant's devices are not suitable for substantial noninfringing use and that defendant is liable as a contributory infringer of the method claims in suit.
In the interest of completeness, we must also determine whether defendant has induced the infringement of the method claims. 35 U.S.C. § 271(b) offers little guidance on when liability for inducing infringement is established:
Whoever actively induces infringement of a patent shall be liable as an infringer.
If defendant were supplying articles or commodities of commerce suitable for substantial noninfringing use, it seems clear that something more than mere knowledge of an intended infringing use would be required to make out a case of active inducement under § 271(b). Federico, Commentary on the New Patent Act, supra, at 53. However, that is not the situation here. Defendant is selling to its distributors and ultimately to their customers articles designed for a particular purpose and to be used in a particular way. And defendant has more than "mere knowledge" that the articles it sells may be used in an infringing manner. Alan Clements testified that he demonstrates "to customers" and "to [his] distributors", Tr. at 2544-45, connected paddle attachments in precisely the manner set forth in the method claims. Thus, defendant sells a product that is obviously designed to function and be used in a specific manner, knows that it is so designed and intends it to be used in an infringing manner, and demonstrates that use to its distributors, presumably so that they may, in turn, do the same for their customers in promoting the sale of the product. The fact that defendant does not actually instruct the ultimate users of the product in its use does not prevent our finding active inducement of infringement, since the intended manner or use of the product is readily apparent. Accordingly, we find defendant liable as a direct infringer of the method claims under 35 U.S.C. § 271(b).
Turning to defendant's second argument, that, even if its acts are sufficient to establish liability as an infringer of the method claims (as we have found), its devices do not meet the language of those claims and thus do not infringe, we find it to be without merit. It is argued that, according to Claims 15 and 17, the separation of the leading attachment is "accomplished" solely by rotation and that since defendant's connections require a pull in addition to rotation to break them, there is no infringement of these claims. This argument is a variant of defendant's previous argument that the word "twist" as used in the article claims meant only rotation. For the same reasons we rejected it there, we reject it here.
IX. Attorneys' Fees
Attorneys' fees may be awarded to the prevailing party in "exceptional cases". 35 U.S.C. § 285. There is nothing exceptional about this case warranting such award. None of the plaintiff's claims were frivolous and none of the defenses were *429 sham. The claims and defenses were asserted and pressed in good faith and were in similar manner vigorously contested. In short, this is not a case where attorney's fees ought to be awarded. Plantronics, Inc. v. Roanwell Corp., 403 F. Supp. 138, 160 (S.D.N.Y.1975), aff'd, 535 F.2d 1397 (2d Cir.), cert. denied, 429 U.S. 1004, 97 S. Ct. 538, 50 L. Ed. 2d 617 (1976).
We hold the claims in suit valid and infringed. The parties shall settle order on ten days' notice.
Appendix to follow.
 Lankton, the patentee, was at the time of his invention operating head of Nylon Products Corporation, a major supplier of plastic attachments to Dennison.
 The patent is reproduced in the Appendix to this opinion.
 1. A clip of a plurality of attached-together attachments each adapted to be engaged with an object, each said attachment comprising an object-penetrating part, an enlarged part, and a narrow elongated section connecting said parts, relatively sturdy frangible means normally operatively connecting said penetrating parts to one another in spaced relationship and adapted to be severed in an attaching device, said elongated sections extending from said penetrating parts, and securing means interposed between and connected to said attachments remote from said frangible means and securing said attachments together, said securing being comparatively more readily severable than said frangible means and adapted to be severed externally of an attaching device.
2. The clip of claim 1, in which said securing means is interposed between and connected to said enlarged parts.
3. The clip of claim 2 in which said securing means is so constructed as to be more readily severable in torsion than in tension.
4. The clip of claim 2 in which said securing means comprises a short, narrow length of material and dimensions such as to have a substantially greater resistance to breaking when tensioned than when twisted.
5. The clip of claim 3 in which said enlarged parts comprise elements arranged essentially parallel to one another with facing surfaces, said securing means being interposed between and secured to said facing surfaces.
7. The clip of claim 4 in which said enlarged parts comprise elements arranged essentially parallel to one another with facing surfaces, said securing means being interposed between and secured to said facing surfaces.
9. The clip of claim 1 in which said securing means is so constructed as to be more readily severable in torsion than in tension.
10. The clip of claim 1 in which said securing means comprises a short, narrow length of material and dimensions such as to have a substantially greater resistance to breaking when tensioned than when twisted.
11. The clip of claim 2 in which said enlarged parts comprise elements arranged essentially parallel to one another with facing surfaces, said securing means being interposed between and secured to said facing surfaces.
14. The method of attaching to an object one of a plurality of attachments which are assembled into a clip by being secured together at first and second points spaced along their length, which method comprises:
a. by means of an attaching device freeing the attachment of said clip at said first point, and engaging said attachment with said object while said attachment remains secured to said clip at said second point, and
b. thereafter, externally of said attaching device, separating said attachment from said clip at said second point.
15. The method of claim 14, in which step (b) is accomplished by rotating said attachment relative to said clip, thereby to exert a substantial torsional force on the means securing said attachment to said clip at said second point.
16. The method of attaching to an object one of a plurality of attachments which are assembled into a clip by being secured together at first and second points spaced along their length, said attachments comprising an object-engaging part, an enlarged part, and a section connecting said parts, said first point being located relatively close to said object-engaging part and said second point being located relatively remote from said object-engaging part, which method comprises:
a. by means of an attaching device freeing the attachment from said clip at said first point, and engaging the object-engaging part of said attachment with said object while said attachment remains secured to said clip at said second point, and
b. thereafter externally of said attaching device separating said attachment from said clip at said second point.
17. The method of claim 16, in which step (b) is accomplished by rotating said attachment relative to said clip, thereby to exert a substantial torsional force on the means securing said attachment to said clip at said second point.
 As it is presumed to be. Western States Mach. Co. v. S. S. Hepworth Co., 147 F.2d 345, 350 (2d Cir.), cert. denied, 325 U.S. 873, 65 S. Ct. 1414, 89 L. Ed. 1991 (1945).
 Our reasons for so concluding will be set forth in greater detail in the section of this opinion dealing with infringement (Section VIII, A, infra).
 Defendant argues (see Defendant's Proposed Finding No. 61) that Claims 14 and 16 are written broadly enough to include within their scope both meanings, and are thus indefinite, but its expert witness, Crews, testified that the "externally of said attaching device" language of Claims 14 and 16 does not describe the devices depicted in Figures 1-13, because those devices are designed to effect the separation of the leading attachment while the clip is still in the gun; and, therefore, the separation does not take place "externally" of the gun. Tr. 2819-23. Crews' suggested interpretation would have these claims read only on the batch embodiment of Col. 8, lines 26-36, in which the T-bar ends of the attachments pierce a series of objects, assembling the objects into a batch from which each object is later separated by severing the relatively strong paddle connection by means of a knife or scissors. To confine Claims 14 and 16 to the batch embodiment is to take too narrow a view of their language. The phrase "externally of said attaching device" is susceptible of both Crews' version and the interpretation that encompasses also those devices depicted in Figures 1-13, in which the connection is broken while the clip remains in the gun, as in Claim 1.
 The "adapted to be severed externally of an attaching device" language is a structural limitation, not merely a description of where the severing is to take place. See In re Venezia, 530 F.2d 956, 958-59 (Cust. & Pat.App.1976).
 Although the so-called "batch embodiment" is described in Col. 8, lines 26-36, the file wrapper indicates that this embodiment was excluded from the article claims by amendment. The applicant stated to the Patent Office that the paddle connection was "relatively flimsy and can readily be manually broken once the attachment has been associated with a tag and garment." Defendant's Exhibit A. Reference to an applicant's amendment to interpret allowed claims in light of others that were cancelled or amended is appropriate. Schriber-Schroth Co. v. Cleveland Trust Co., 311 U.S. 211, 220-21, 61 S. Ct. 235, 85 L. Ed. 132 (1940).
 The only item of prior art that was not cited to the patent examiner that might arguably be more relevant than those items which were considered is Japanese Utility Model Publication 45/XXXX-XXXXX, discussed below. However, we think its relevance is significantly undermined by its unworkability and lack of clear teachings. In any event, its relevance to the claims of the patent in suit does not approach the degree of relevance to the claims in suit of the prior art in Julie, supra (individual features of patent had been "predicted, known, and actually used," 501 F.2d at 1133, and had a "direct predecessor," 501 F.2d at 1134, in the prior art, and their combination was also obvious); Reeves, supra (omitted prior art directly countered a primary contention of patentee that his process was unobvious to one skilled in the art); Lerner, supra (omitted prior art disclosed feature whose significance was emphasized during prosecution of the patent in suit, 406 F.Supp. at 563); and Kahn, supra (patentee omitted reference to paper disclosing many elements of invention and misstated teachings of prior patent which disclosed feature "essentially similar", 367 F. Supp. at 71, to salient feature of patentee's invention). Thus, even if this reference were considered to be more relevant than the prior art cited to the patent examiner, its effect on the presumption of validity would be minimal.
 Epstein v. Dennison, 314 F. Supp. 116 (S.D. N.Y.1969).
 A machine used in parimutuel betting.
 Kirk teaches that the stretching is performed by separating the ends of the mold before the devices are removed from the mold. This is referred to as "in-the-mold" stretching, as opposed to stretching which takes place outside of the mold, as an independent operation.
 Powers discloses that a variety of materials may be used and specifies "at least one" connection between each fastener. Spinney's connections are not necessarily thin and easily severable they are intended to be severed at each end of the connection by a cutting element in a machine.
 Patents pending at the time of application are relevant prior art. Hazeltine Research, Inc. v. Brenner, 382 U.S. 252, 86 S. Ct. 335, 15 L. Ed. 2d 304 (1965); Reeves Brothers, Inc. v. U. S. Laminating Corp., supra, 417 F.2d at 871 n.4.
 Defendant has, thankfully, provided the Court with a translation of JUMP 45/XXXX-XXXXX. Defendant's Exhibit D-2.
 A printed publication in a foreign country is part of the prior art. 35 U.S.C. § 102(b).
 "Gating", for our purposes, may be described as providing connections between a group of items which are molded in a single molding and the frame which holds the items in place for further processing and handling. The connection is designed to be strong enough to withstand whatever dislocative forces are normally applied but weak enough to allow severance when the item is to be used. A common example of plastic "gates" is the connection holding plastic model parts to a frame.
 Defendant's argument that because JUMP is silent on the relative strengths of the connections, it cannot be assumed that they are of the same strength, is specious. There is nothing in that publication that suggests that the connections may be of different strengths.
 Plaintiff argues that JUMP fails to disclose adhesive between the paddles, thus differing from Claims 2 and 11 of the patent in suit. This difference is insignificant, one which would surely be obvious to a lay-person, not to mention one skilled in the art.
 We compare the claims in suit to these three publications because the other patents cited above add nothing of relevance to Schumm, Keck and JUMP. We do not imply that the latter two publications are more pertinent than the prior art cited to the examiner (see our discussion of the presumption of validity, supra).
 Although an unworkable prior patent or publication is part of the prior art, David & David, Inc. v. Myerson, 388 F.2d 292, 295 n.2 (2d Cir. 1968); Ling-Temco-Vought, Inc. v. Kollsman Instrument Corp., 372 F.2d 263, 267 (2d Cir. 1967), its unworkability bears on what it teaches and on what would be obvious from it. Leach v. Rockwood & Co., 273 F. Supp. 779, 789 (W.D.Wis.1967), aff'd, 404 F.2d 652 (7th Cir. 1968).
 Although the connected paddle devices are cheaper to manufacture than the unconnected, plaintiff chooses to meet price competition with the unconnecteds in order to maintain the connecteds as its premium product.
 Plaintiff originally did not manufacture its fine fastener with connections because it thought that the pull necessary to break the connections would damage the fine, delicate fabrics with which the fine fasteners were designed to be used. Tr. at 342.
 The difference between December, 1971 and April, 1972 is not of great significance in this case, where plaintiff has made no claim that the Hi-Lock was stolen from Dennison. The situation might be different if such a claim were made, since Lankton's conception allegedly took place in February of 1972.
 Only the April 28th letter specifies connected paddles.
 The apparent lack of commercial success of the Hi-Lock device is further support for our finding that the flash connection was not intended, as it indicates that the connection is not a benefit but a detriment in the Hi-Lock's intended use, that of forming a loop. The lack of success of the Hi-Lock is also evidence that the uses demonstrated by defendant, which we refer to as "contrived", were not the primary uses or even common uses of the Hi-Lock. If they were, there does not appear to be any reason why the Hi-Lock would not have enjoyed the success of the device described in the claims in suit. In its intended use, the connections are clearly a detriment.
 Plaintiff concedes that it knows of no instance in which defendant saw its attachments during the testing period or before and that it has no direct proof of appropriation.
 It is quite possible and not uncommon for two or more solutions to a problem to emerge at substantially the same time without either of them being obvious. See 35 U.S.C. § 102(g).
 35 U.S.C. § 271 reads in pertinent part:
(a) Except as otherwise provided in this title, whoever without authority makes, uses or sells any patented invention, within the United States during the term of the patent therefor, infringes the patent.
 We considered the validity of these omitted claims, supra, because defendant asserted as a counterclaim their invalidity, as to which we felt defendant was entitled to a judgment. However, we do not feel similarly compelled with respect to whether these claims are infringed, particularly in light of the fact that the proof at trial, as far as we can tell, did not relate specifically to these claims.
 Although we need not decide the question, even if we were to limit Claims 1, 2 and 11 to the teaching of Col. 6, lines 18-36, we would nevertheless hold them to be infringed under the doctrine of equivalents. Defendant's structures employ substantially the same means and operate in substantially the same manner to accomplish substantially the same result, Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 608, 70 S. Ct. 854, 94 L. Ed. 1097 (1950); N. V. Maatschappij Voor Industriele Waarden v. A. O. Smith Corp., 590 F.2d 415, 421 (2d Cir. 1978); Kaz Mfg. Co. v. Northern Electric Co., 412 F. Supp. 470, 476 (S.D.N.Y. 1976) (Conner, J.), and plaintiff is not estopped to charge infringement of these claims by defendant's structures.
 That "twist" was meant to include rotation and pull is further evidenced by the recitation found in Col. 8, lines 16-25, that "the existence of the securing means . . . will have the effect of checking whether the leading attachment has been secured to the object . .if it has not, there will not be sufficient resistance to movement on the part of that attachment, the securing means . . . will not break, and the leading attachment will be pulled away from the object . . . along with the attaching gun . . . and the remainder of the clip. . . ."
 As with Claims 1, 2 and 11, supra, even if we were to find that "twist" as used in Claims 4 and 10 was intended to mean rotation alone, we would nonetheless find these claims infringed under the doctrine of equivalents. Plaintiff is not estopped to claim that "twist" includes a pulling motion.
 The parties have not cited, and our own research has not found, any case directly on point; that is, where a patented device has been supplied, in itself an act of direct infringement of the article claims of a patent, to someone whose use thereof may constitute a direct infringement of method claims contained in the same patent.
 It is interesting to note that one of these occasions was in the marking room of a large customer that had ordered unconnected paddles but had been shipped the connected version in error. Tr. at 3276.