Appeal from a judgment of the United States District Court for the Northern District of Illinois, Eastern Division, holding claims 1-4, 7, 10, 12, 14-22 and 24-27 of United States Patent No. 3,537,146 ('146 patent), claims 1, 10 and 12 of United States Patent No. 3,660,869 ('869 patent), and claims 1, 2, 5, 6, 11, 17, and 21 of United States Patent No. 3,965,538 ('538 patent) invalid on the ground of obviousness, and the same claims of the '538 patent invalid on the ground of double patenting. We reverse.

(1) The Art

One-piece cable ties are used to bind a bundle of cables or insulated wires. Looped around the bundle, a strap has one end passed through an opening in a frame at the other end. Teeth on the strap engage with a locking device at the frame. The strap and device operate to permit tensioning of the strap end to prevent its loosening or withdrawal.

Prior art cable ties with rigid locking devices had a desired high strap withdrawal force, but an undesired high strap insertion force. Conversely, those with flexible locking devices had an undesired low withdrawal force and a desired low insertion force.

(2) The Real World Story

As stated in Rosemount, Inc. v. Beckman Instruments, Inc., 727 F.2d 1540, 1544, 221 U.S.P.Q. (BNA) 1, 5 (Fed.Cir.1984), many patent suits "arise out of the affairs of people, real people facing real problems." That is true of the present suit.

Jack E. Caveney founded Panduit in the basement of his home in 1955, making first a plastic wiring duct. In 1958, he developed and began selling a two-piece plastic cable tie. In 1961, he began a research program to develop a one-piece plastic cable tie. That program lasted nine years and cost several million dollars. The end result of that program is the tie of the '538 patent, which includes the features of all three patents in suit. Primary among its advantages are the minimal force required to deflect the locking member when the strap is inserted through the frame and the high force required to withdraw the strap from the frame. Jack Caveney was the first to envision and achieve a cable tie having a higher ratio of low strap insertion force and high strap withdrawal force than anyone else had ever achieved or thought possible. Panduit's commercial embodiment of the patents in suit achieves an insertion force of one-half pound and a withdrawal force of 80 pounds.

First sold in 1970, the tie of the '538 patent had by 1984 achieved annual sales of $50 million, and was accounting for half of Panduit's total profits and 80 percent of its cable tie sales.

Beginning in 1968, Dennison Manufacturing Corporation (Dennison) put its staff of engineers and designers to work on a one-piece cable tie development program. It carried that program on at considerable expense for ten years, developing many ties and patenting some. None was successful. With the '869 patent before it, Dennison copied the tie claimed in that patent in 1976. When the '538 patent issued in 1976, Dennison thereupon copied the tie claimed in that patent. Having failed to succeed for over ten years with ties of its own design, Dennison achieved such success with its copy of the '538 tie as to make Dennison the second or third largest supplier of one-piece cable ties.

(3) The Patents in Suit1

The '146 patent issued November 3, 1970 on an application filed August 6, 1968. Figure 7 is representative:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The '869 patent issued May 9, 1972 on an application filed May 1, 1969. Figure 9 is representative:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The '538 patent issued June 29, 1976 on an application filed May 5, 1969. Figure 9 is representative:

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

(4) InfringementDennison admits that the results of its research and development program, its Merser 8 and Merser 9 ties, "are covered by the representative claims of the patents-in-suit." Dennison's Merser 8 tie is admittedly covered by the asserted claims of the '146 and '869 patents. Dennison's Merser 9A tie is admittedly covered by the asserted claims of all three patents in suit. Dennison's defense was, accordingly, limited to its asserted invalidity of the patents in suit.

(5) Proceedings before the District Court

On December 14, 1978, Panduit sued Dennison, alleging infringement of certain claims of its '146 and '869 patents. Dennison answered on February 16, 1979, asserting invalidity of those patents. In 1981, Panduit amended its complaint, alleging infringement of certain claims of the '538 patent. Dennison's answer asserted invalidity of that patent.

After years of skirmish, trial was held on 13 dates between October 9 and November 8, 1984. A judgment was entered on December 4, 1984, following the court's ruling from the bench in favor of Dennison.

(6) The District Court's Decision

(a) Sec. 103

(I) The '146 Patent

The district court said the closest prior art reference was United States Patent No. 3,127,648 to Emery (Emery). Panduit had cited Emery to the Examiner, who considered it during prosecution. As evaluated by the district court, however, Emery "had a problem," i.e., "that the strap of the cable tie protruded from the head2 when the cable tie was in tension," and because "there was only one tooth on the Emery tie ... any pulling of the protruding end of the strap could easily disengage that one tooth."

When Panduit at trial said the claimed subject matter was patentably distinct in view of plural teeth on the pawl, sequential engagement of the pawl teeth with those on the strap, and wedging of those engaged members against the abutting wall, the district court responded that what it called Panduit's "new elements" were "simply a description of one element, not three," saying: "it seems to me that one element has been broken down into three elements, and in so doing [Panduit] is able to argue that it has more patentable differences over Emery than I find to be the case."3

In response to Panduit's argument that Emery did not wedge the strap against the abutment wall, the district court said that "the text of the Emery patent makes abundantly clear that Emery's intention was to achieve wedging against the abutment wall," and that such wedging "can only be accomplished if [it] is inside the head, because that is where the abutting wall is."4

When Panduit pointed out that Emery's drawing clearly shows engagement of the pawl and strap outside the frame, the district court said:

Now, how do we account for the drawing? I can't account for that drawing ... It's an aberration, it's an anomaly. And as I indicated a little earlier this afternoon, it is difficult to believe that someone whose only task was to make two drawings, could make a mistake on one of them when his business is making patent drawings. And yet, I adopt defendant's suggestion that what is represented by Figure 2 of Emery is a draftsman's error.5

The district court concluded that the solution to Emery's "release problem" was "to add more teeth and put those teeth exactly where the intelligent reader of Emery would have known they should be, namely on the pawl, inside the head, gripping the strap against the abutment wall." The district court concluded that claim 24 was "necessarily implied in a decision to use multiple teeth," and that use of multiple teeth "was, nonetheless, obvious in light of the problem presented in Emery." To support that determination, the district court looked to two references cited during prosecution, each of which disclosed multiple teeth, concluding: "I don't even think it is necessary for anybody to teach [multiple teeth] in view of the fact that it was an obvious solution."

(II) The '869 Patent

The district court said, "the problem addressed by [the '869 patent] was insufficient strength of compression." The court attributed that "problem" to "the hinge"6 being positioned "essentially perpendicular to the axis of the strap." The district court said "[t]he solution devised by Caveney was to change the orientation of the pawl to a more or less vertical position, so that it would be essentially parallel to the axis of the tensioned strap, and then to place a ledge under the pawl to absorb the compressive force."

The district court found that "to increase the compressive strength of the cable tie, and specifically the pawl and hinge of the cable tie, you would set the pawl on a ledge." It further found that "it is obvious that a ledge should be an integral part of the head of the tie," and that "extending the ledge from side wall to side wall is an obvious way of accomplishing that."

The district court also looked to three references: United States Patent No. 3,484,905 to Eberhardt (Eberhardt), filed over three months after the '869 patent application (as a "contemporaneous development"); United States Patent No. 3,214,808 to Litwin (Litwin); and Japanese Utility Model Publication No. 27 023/1964 (Japanese patent), the first two of which had been cited during prosecution. The court said those references "showed structures analogous to the '869 ledge in that they absorbed compressive force."

The court concluded that "the result achieved by introducing a ledge in the '869 patent was altogether expected," saying that "where the result of a rather ordinary step is exactly what you would expect ... that is an argument for obviousness," and that the claimed invention of the '869 patent would "be obvious7 both on general principles of physics and in light of the Japanese patent and Litwin patent."

(III) The '538 Patent

The district court said the "principal problem" addressed in the '538 patent was that the pawl of the '869 patent was not sufficiently flexible, and that its teeth tended to break off when tensioned in a "file like action". This it termed the "rigidity" problem, saying:

Now, if the problem is rigidity, the opposite of rigidity is flexibility. So it seems to me that the person of ordinary skill in the art, when recognizing the problem--and that, incidentally, would not have been difficult--there is testimony in the record that the problem was obvious, the problem itself was obvious--the teeth were being filed off--the person of ordinary skill in the art would necessarily think in terms of making the pawl more flexible than it was. The objective would be to achieve some way of enabling the pawl to get out of the way of the strap as it moved past so that the teeth would not be filed off.

The district court said the '538 patent avoided the rigidity problem by using what it called a "discrete hinge," which it in turn called the "essence of the '538 patent." That "solution" was necessary, it believed, because only a discrete hinge could achieve flexibility while not interfering with multiple engagement of the teeth.

The district court pointed to what it termed "hinges and discrete hinges" in the Emery, Litwin, and Fein patents, and to United States Patent No. 3,486,201 to Bourne (filed three months after the '538 patent application) (as a "contemporaneous development").8

The district court also said the '146 patent showed a "discrete hinge" and thus was "perhaps the most relevant prior art".9

The district court said Panduit claimed the discrete hinge in the '538 patent in language that "described a number of necessary characteristics of the same element as though they were separate elements."10

The primary prior art basis for the court's holding obvious the claimed inventions of the '538 patent was that the '538 patent offered "nothing ... other than a discrete hinge, that is new over the '869 patent."11

(IV) Objective Evidence

Respecting the objective evidence (the so-called "secondary considerations"), the district court said: "Mr. Cavaney and Mr. Moody and Panduit have designed and perfected an excellent and, indeed, preeminent commercial product. And the fact that they have done so contributed greatly to the difficulty of this case. It is not easy to find something obvious in the face of those facts. And I do so with great reluctance and great resistance."

(V) Concluding Remarks Respecting Sec. 103

Summarizing, the district court described Panduit's approach as "flawed by two errors".

First, it said Panduit's claims split "what are essentially the same element into a number of parts,"12 and presented "confusing claims to the Patent Office over a long series of patent applications for various cable ties," accompanied by "inconsistent and self-contradictory arguments."13

Second, it said that Panduit overly relied on the absence from the prior art of anything even close to the claimed inventions, saying Panduit gave "insufficient attention to what is taught by general engineering principles and general principles of physics and, indeed, the common experience of mankind."14

(b) Section 102(g)

Respecting Dennison's allegation of invalidity of the '146 patent under 35 U.S.C. Sec. 102(g), the district court determined that, in reducing to practice the subject matter of the '146 patent in 1961 and filing an application in 1968, Panduit had not "forfeited any of its patents," that Dennison failed to show "prejudice, reliance, or some change in position," and that "mere delay [in filing] without more does not result in forfeiture."

(c) Double Patenting

In its statement, the district court upheld this defense in two sentences:

Turning to the question of double patenting, I find that there is double patenting. The reason for this is that the discrete hinge in the '538 patent is not a patentable difference over the '869 patent.

(d) Copying

The district court found that Dennison copied the claimed inventions of the Panduit patents, saying "[i]f there were any doubts on that, it would have been dispelled by Dennison's failure to bring to court the people who could have denied that, if it were possible to deny it." Instead, it said, Dennison's witnesses had "secondhand and relatively remote connection with the accused devices," and their testimony was "of very little assistance."

(e) Attorney Fees

The court denied Dennison's demand for attorney fees, saying "this in not an exceptional case within the meaning of the statute." It viewed Panduit's suit neither as frivolous nor brought in bad faith, and noted that "there is room for some difference of opinion here based largely upon those secondary considerations that have given me such trouble in the case: failure by others, copying by Dennison, and unquestioned commercial success of the plaintiff."

(I) Whether the district court erred in holding the claims in suit invalid for obviousness.

(II) Whether the district court erred in holding the claims of '538 patent invalid for double patenting.

(III) Whether the district court erred in refusing to hold the claims of the '146 patent invalid under 35 U.S.C. Sec. 102(g).

(I) Obviousness

(a) The Landscape

The present record reflects the insidious and powerful phenomenon known in patent law as the use of hindsight, for in this case a most careful and conscientious judge, after voicing the caveat against it, was nonetheless victimized by that phenomenon.

That counsel may succeed in leading a court to disregard the statutory requirement for consideration of the claimed invention "as a whole", to look for its "essence" in a single feature, and to pick and choose features found somewhere in the prior art, though the court had recognized the impropriety of that decisional approach, is a second phenomenon forming part of the landscape of the present case.

That a court may misconstrue its role as requiring it to decide whether it considers an invention patentable, in place of the statutory requirement that it decide whether the patent's challenger has carried its burden of proving by clear and convincing evidence the facts compelling a conclusion of patent invalidity, SSIH Equipment, S.A. v. U.S. Intern. Trade Comm'n, 718 F.2d 365, 375, 218 U.S.P.Q. (BNA) 678, 687 (Fed.Cir.1983), is a third phenomenon in the landscape of this case.

That a court may be led to concentrate on lawyers' games played with the patent in suit and those in the prior art, while allowing an obscuration of the real world story that forms a major part of the landscape of this case, is a fourth phenomenon here present.

At the outset of its statement from the bench, the district court reflected a clear instinct for the proper parameters of decisionmaking under 35 U.S.C. Sec. 103:

It is not appropriate for the Court to engage in hindsight. Furthermore, it is not appropriate for the Court to pick and choose isolated elements from various prior art references and combine them so as to yield the invention in question when such combining would not have been an obvious thing to do at the time in question.

* * *

* * *

I am satisfied that in plaintiff's patents we see the best of the art at the times in question. The '146 patent was better than anything else that then existed and the second two patents, the '896 and '538 patents, were that best at the time.

In fact, [Panduit's] commercial embodiment of the '538 patent appears to be the leader in the industry. As I have remarked at various times during the trial, the evidence is simply overwhelming that Dennison copied the '538 patent.

Now, this commercial success, the failure of others and the copying by Dennison argue strongly in favor of non-obviousness. In addition, of course, [Panduit] has the benefit of the rule that the presumption of validity of the patent can be overcome only by clear and convincing evidence.

It is with these caveats firmly in mind that I will now undertake an analysis of each patent.

There was an eleven day recess in the closing arguments. Before the recess, the district court had said, "obviously no previous patent showed the combination at issue here. I am not persuaded that this invention is similar to any one piece of prior art," and "other people were going in different ways and did not arrive at the Caveney structure or anything substantially identical to it."

The present case establishes that, even on a journey to decision so well begun, a busy court may lose its way. Speculation on just which fork in the road led to error is fruitless. There were, moreover, numerous forks and numerous errors.15

(b) The Errors

En route to its decision on obviousness, the district court was led to these errors:

(1) It employed the benefit of hindsight.

(2) It misinterpreted the claimed inventions.

(3) It misevaluated the prior art.

(4) It misconstrued its role.

(5) It applied an improper and impossible standard of obviousness.

(6) It gave too little weight to the objective evidence of non-obviousness and the real world story reflected in that evidence.

(1) Hindsight

The record compels the conclusion that the district court, having heard many days of testimony from the inventor Caveney, was unable to cast the mind back to the "time the invention was made." 35 U.S.C. Sec. 103. The court did not, as the statute requires, view the prior art from the perspective of one skilled in the art and uninformed by that testimony. Caveney described in detail the shortcomings of prior art cable ties and the superiority of cable ties constructed in accord with the claims in suit. In deciding the obviousness question, the district court looked to knowledge taught by the inventor Caveney, in his patents and in his testimony, and then used that knowledge against its teacher.

The test is whether the subject matter of the claimed inventions would have been obvious to one skilled in the art at the time the inventions were made, not what would be obvious to a judge after reading the patents in suit and hearing the testimony. In the present case, for example, there is no way that one skilled in the art in 1961 and 1968 (necessarily unaided by knowledge of the patents in suit and Caveney's testimony) would find in the prior art either a teaching or a suggestion of the claimed inventions.

Though hindsight was applied to all of the claims and references, a few examples are illustrative. There is no way one skilled in the art would have focused on an incomprehensible phrase in Fein, and ignored his disclosure of a rigid pawl; or would have focused on multiple teeth in Fein and Litwin, while ignoring how those reference structures function and the total absence of an abutment wall; or would have known that the teeth must be "in a straight line and equidistant from the surface of the abutment wall"; or would have learned from the Japanese and Litwin patents, which disclose manually releasable pawls, how to design the non-releasable pawls of the claimed inventions. W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1550, 220 U.S.P.Q. (BNA) 303, 311 (Fed.Cir.1983). Most importantly, nothing whatever in the prior art would have suggested to one skilled in the art how to produce a cable tie with a one-half pound insertion force and an 80 pound withdrawal force.16

The use of hindsight was the more egregious here, where, as discussed below, the court applied a subjective view based on "engineering principles and general principles of physics" and "the common experience of mankind" in the face of a record establishing that those "principles" and that "experience" did not for years suggest the claimed inventions to numerous earlier workers skilled in the art, including Dennison's engineers.

This court and one of its predecessors have cautioned against the use of hindsight:

It is difficult but necessary that the decisionmaker forget what he or she has been taught at trial about the claimed invention and cast the mind back to the time the invention was made (often as here many years), to occupy the mind of one skilled in the art who is presented only with the references, and who is normally guided by the then-accepted wisdom in the art.

W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d at 1553, 220 USPQ at 313.

It is impermissible to first ascertain factually what appellants did and then view the prior art in such a manner as to select from the random facts of that art only those which may be modified and then utilized to reconstruct appellants' invention from such prior art.

In re Shuman, 361 F.2d 1008, 1012, 150 U.S.P.Q. (BNA) 54, 57 (CCPA 1966).

In Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1538, 218 U.S.P.Q. (BNA) 871, 879 (Fed.Cir.1983), this court noted the conflict with the statute, 35 U.S.C. Sec. 103, when the test applied is whether the claimed invention would have been obvious, not to one skilled in the art when the invention was made, but "to a judge or other layman after learning all about the invention."17

(2) Misinterpretation of the Claimed Inventions

The statute requires that the subject matter of the claimed inventions be considered "as a whole". 35 U.S.C. Sec. 103. In the present case, the district court interpreted the claims of the '146 patent as though they were drawn to "multiple teeth", those of the '896 patent as though they were drawn to a "ledge", and those of the '538 patent as though they were drawn to a "hinge". That was error.18

The court was apparently led into the instant error by defense counsel's contention that each patent should be evaluated for what counsel called its "essence". In so arguing, counsel misled the court in respect of the law when he told the court that if it upheld the validity of the patents it would be telling the entire industry:

"You can't have more than one parallel tooth." (re the '146 patent);

"You can't mount a pawl on a ledge." (re the '869 patent);

"You can't use a neck in there when you support your pawl." (re the '538 patent).

Counsel knew, or should have known, of the falsity of the foregoing statements. Those statements mistate the law of validity and the law of infringement. As above indicated, validity is determined on the basis of the claimed subject matter as a whole, 35 U.S.C. Sec. 103, not in respect of a single element. Similarly, infringement of a claimed combination requires the presence in an accused structure of each claimed element, or its equivalent. Upholding of the patents in suit does not, therefore, preclude any worker in the art from employing multiple teeth, ledges, and hinges in whatever combinations the worker may desire, so long as those combinations are distinct from those claimed in the present patents.

This court has cautioned against the claim dissection that occurred here:

Though it is proper to note the difference in a claimed invention from the prior art ... it is improper ... to consider the difference as the invention. The 'difference' may have seemed slight (as has been the case with some of history's great inventions, e.g. the telephone), but it may have been the key to success and advancement in the art resulting from the invention.

* * *

* * *

Hence the statute, the law established not by judges but by Congress, requires that the invention as claimed be considered "as a whole" when considering whether that invention would have been obvious when it was made. 35 U.S.C. Sec. 103.

Jones v. Hardy, 727 F.2d 1524, 1528, 220 U.S.P.Q. (BNA) 1021, 1024 (Fed.Cir.1984).

(3) Misevaluation of the Prior Art

The well established rule of law is that each prior art reference must be evaluated as an entirety, and that all of the prior art must beevaluated as a whole. See W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d at 1550, 220 USPQ at 311; In re Kuderna, 426 F.2d 385, 390, 165 U.S.P.Q. (BNA) 575, 578-79 (CCPA 1970). The rule, as rules should, cuts both ways.

Inventions have often been held to have been obvious where no single reference disclosed the claimed invention, and in the face of arguments that the reference structures could not practically be "combined". That result was compelled because the art as a whole in those cases suggested the combination claimed. See, e.g., In re Yamamoto, 740 F.2d 1569, 1573, 222 U.S.P.Q. (BNA) 934, 937 (Fed.Cir.1984); Orthopedic Equipment v. United States, 702 F.2d 1005, 1013, 217 U.S.P.Q. (BNA) 193, 200 (Fed.Cir.1983).

At the same time, inventions have been held to have been nonobvious where neither any reference, considered in its entirety, nor the prior art as a whole, suggested the combination claimed. See, e.g., Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549, 1556, 225 U.S.P.Q. (BNA) 26, 31 (Fed.Cir.1985); ACS Hospital Systems, Inc. v. Montefiore Hospital, 732 F.2d 1572, 1577, 221 U.S.P.Q. (BNA) 929, 932-33 (Fed.Cir.1984). That is the circumstance of record here.

Nowhere did the district court indicate where in the prior art there might be a suggestion for combining teachings of individual references, or how, if there were such suggestion, such a combination would equal any of those claimed in the patents in suit. There was here simply nothing in the prior art to rebut the presumption of validity to which the patents in suit are statutorily entitled.

What happened here occurred in W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d at 1550, 1552, 221 U.S.P.Q. (BNA) at 311, 312, where this court stated:

In its consideration of the prior art, however, the district court erred ... in considering the claims in less than their entireties ... and in considering the references in less than their entireties, i.e., in disregarding disclosures in the references that diverge from and teach away from the invention at hand....

* * *

* * *

... The result is that the claims were used as a frame, and individual naked parts of separate prior art references were employed as a mosaic to recreate a facsimile of the claimed invention.

As indicated in notes 2-5, and 7-10, supra, the court made specific technical and legal errors in evaluating the individual prior art patents, none of which was even close to the claimed inventions. The court made a basic error under the influence of hindsight, when it found in the prior art "problems" which were not extant in that prior art, but which were problems inventor Caveney had seen and solved and told the court about. The errors cited in the foregoing notes need not be further discussed here.

Three fundamental errors resulted from a disregard of the decisional parameters governing the proper evaluation of prior art. The first, as above indicated, was the picking and choosing of "teeth", "ledges", and "hinges" either absent from the prior art references or there disclosed in entirely distinct form, characteristics, and relationships. It must be remembered that the Examiner is required to consider references in their entireties, i.e., including those portions that would argue against obviousness. The statute governing the obvious/non-obvious determination in light of the prior art, 35 U.S.C. Sec. 103, is as applicable to courts as it is to Examiners. Courts are not, of course, bound by the Examiner's determination, but a court's route to a contrary determination is governed by the same statute, and must not be based on a superficial review of the prior art like that here employed.

The second fundamental error was a disregard of the Examiner's consideration of all but one of the references relied on by Dennison, and of the irrelevancy of that one reference (Japanese patent, disclosing a releasable pawl). Nowhere does the district court indicate any basis for its disagreement with the Examiner's evaluation of the references, an oversight that may have been prompted by defense counsel's statement to the court: "The fact that the Patent Office [sic] allowed a claim is little evidence of whether there is anything patentable in the claim." Under the statute, 35 U.S.C. Sec. 282, the fact of allowance is at least sufficient "evidence" to place on the challenger the burden of proving the Examiner wrong.

The third fundamental error was the refusal of the court to credit the real world environment surrounding the inventions disclosed in the reference patents and in the patents in suit. Referring to the reference structures, the court paid no attention to insertion and withdrawal forces in the reference structures, or to the striking differences in the references, treating them as equivalents of each other. The court specifically stated, "I don't need to know all the details about how all these other things work or don't work." But "how things work" is critical to encouragement of every research and development activity, and every advancement of the arts useful to the public, the very purposes of the patent system.

It was precisely "how things work" that made the prior art cable ties failures and the cable tie of the patent in suit the most successful in the history of the industry. It was "how things work" that caused Dennison's ten year development to fail and led Dennison to copy the tie of the patents in suit. It was "how things work" that enabled Caveney alone to achieve a one-half to 80 pound ratio between insertion and withdrawal forces. And it was a refusal to consider "how things work" that caused the court to cite isolated minutiae from various references, while ignoring critically important structural distinctions that significantly affect the different achievements of which the reference and claimed structures are capable.19

The court appears to have been led, at least in part, to its erroneous evaluation of the prior art by defense counsel's trial tactic. Working at a large easel, counsel made free-hand sketches. Beginning with a sketch purported to represent a prior art structure (but with no claimed feature), counsel added a ledge, or a hinge, or a frame, and in minutes produced a cable tie design having all the features set forth in a claim. Counsel then offered his completed sketches as exhibits. They were admitted over Panduit's "strong objection". The court's statement from the bench indicates that it substituted counsel's sketches for the prior art patents, and that it was impressed with the ease and speed with which counsel created the claimed inventions in the courtroom. That procedure was inappropriate.

Counsel was not under oath and not subject to cross-examination. Admission of his "testimony" in the form of his sketches was non-probative. As above indicated, prior art references must be evaluated on what they taught or suggested in their entireties when the invention was made, not on hypothetical modifications made with knowledge of the invention in suit many years later. That rule is applicable to witnesses and, in even greater force, to counsel's "testimony". There was simply no justification for the conjectural modifications reflected in counsel's sketches. See Carl Schenck, A.G. v. Nortron Corp., 713 F.2d 782, 787, 218 U.S.P.Q. (BNA) 698, 702 (Fed.Cir.1983) ("modification unwarranted by the disclosure of a reference is improper").

Dennison does not rely on its counsel-prepared exhibits before this court. That is well, for the record is compelling that the inventing of cable ties in not the simple matter reflected in the trial tactic of defense counsel. Panduit invested nine years and several million dollars before achieving the successful tie of the patents in suit. Dennison tried for ten years, at a presumably similar expense, and failed. Dennison's prior art book contained twelve patents on pre-Caveney cable ties that were either never sold or were hopelessly outclassed in the marketplace by the cable tie of the patents in suit.

Indeed, it was only the district court who believed that cable tie inventing was a simple, obvious, routine matter of finding elements in the prior art and modifying them in accord with "principles of physics" and "common experience". To reach that belief, the court had to ignore what workers of ordinary skill did in the real world with those same principles and that same experience over the years (as shown in the whole of the prior art references). It also had to ignore the testimony of Dennison's own deposition witnesses. Dennison's chief engineer, Jack Stewart, said "the design of these ties is not as simple as it may appear on the surface." Stewart also said Dennison's research program was "tantalizingly close to success on many occasions." Dennison's senior engineer, Joyce, said "a good tie is a damn difficult thing to design, extremely difficult."20

Properly evaluated, the prior art in this case confirms the non-obviousness of the claimed inventions in suit. The misevaluation applied here was of such nature as to destroy any patent. It was plain error.

(4) The Role of the Court

Patents are born valid. 35 U.S.C. Sec. 282; Roper Corp. v. Litton Systems, Inc., 757 F.2d 1266, 1270, 225 U.S.P.Q. (BNA) 345, 347 (Fed.Cir.1985). It is not, therefore, a required role of a court, in a suit for patent infringement, to declare patents valid.21 Nor is it the court's role to start from scratch, as a surrogate Examiner, to referee de novo a dispute on the validity question.

Though that procedure would render the Patent and Trademark Office (PTO) unnecessary, it was essentially followed here. As above indicated, courts are clearly not bound by the events in the PTO or by its decisions. Fromson v. Advance Offset Plate, Inc., 755 F.2d at 1555, 225 USPQ at 31. On the other hand, there is no point or value in, and no warrant for, proceeding as though nothing happened before suit except the naked issue of a patent.

The statute mandating a presumption of validity, 35 U.S.C. Sec. 282, places the burden of proving facts compelling a conclusion of invalidity on the party asserting invalidity. This court has said the burden of proving facts compelling a conclusion of patent invalidity must be carried by clear and convincing evidence. SSIH Equipment, S.A. v. U.S. Intern. Trade Comm'n, 718 F.2d 365, 375, 218 U.S.P.Q. (BNA) 678, 687 (Fed.Cir.1983). The role of the court is to determine whether the validity challenger carried that burden. If the court determines that that burden was not carried, it need only so state. If the court determines that that burden was carried, it should declare the patent invalid. See Lindemann Maschinenfabrick GMBH v. American Hoist & Derrick Co., 730 F.2d 1452, 1452 n. 1, 221 U.S.P.Q. (BNA) 481, 484 n. 1 (Fed.Cir.1984).

Though the court here referred at the outset to the presumption of validity, mere lip service is insufficient. If the statute is not to be nullified, courts and lawyers must render it more than a mere nod in passing. It must be applied. Counsel for the patent should not undertake the non-existent initial burden of proving validity, but must be prepared to rebut whatever case of invalidity may be made by the patent's challenger. If counsel so misperceive their roles as to indicate a reversal, it should not be surprising that a court may be led to misperceive its own.

In the present case, as indicated above, n. 14, the district court said it had been "going back and forth" on the question of obviousness. It recognized the possible decisional influence of an overwhelming record of objective evidence (commercial success, failure of others, others going a different way, copying by Dennison), saying "this is a case that raises that question about as squarely as any I've encountered." It praised the "patents" [sic, claimed cable ties] as "better than anything else that then existed" and "best at the time." It said "there is room for some difference of opinion here based largely on those secondary considerations that have given me so much trouble in the case." Noting the cable tie of the patents in suit was an "excellent and, indeed preeminent commercial product," the court said it was "not easy" to reach its obviousness conclusion, but "I do so with great reluctance and great resistance." The court's uncertainty should have, in view of Sec. 282, ended the obviousness inquiry.

It is at best difficult to find in the words of the court a firm conviction that Dennison had carried its burden of establishing facts supporting invalidity of the patents in suit by clear and convincing evidence. On the contrary, the court indicated that it was not considering that question, but was considering whether it thought the claimed inventions obvious, and was, moreover, in substantial doubt on even that score.

Thus the district court gave too little, if any, effect to the presumption of validity, 35 U.S.C. Sec. 282. That presumption is procedural, placing the burden of its rebuttal on the patent challenger. The rebuttal here presented by Dennison was based on prior art references fully considered by the Examiner who examined all three patents in suit, and a cumulative reference. As above indicated, those references, when properly evaluated, were, individually and collectively, totally inadequate. A proper evaluation of the prior art would therefore have established the failure of Dennison to carry its statutory burden and would have ended the obviousness inquiry in this case.

(5) Improper and Impossible Standards

(a) General Principles and Common Experience

As above indicated, the district court in reaching its obviousness conclusion relied heavily on what it thought was "taught by general engineering principles and general principles of physics and, indeed, the common experience of mankind."22 That approach is error for a number of reasons.

First, it is in conflict with the statute. Section 103 says nothing of such "principles" and "experience", but refers to whether the subject matter of inventions not anticipated by something in the "prior art" would nonetheless have been obvious in view of that art.

Second, it is in conflict with the patent examining system, which is based on examination of application claims in light of the prior art.

Third, it raises a standard impossible for any patent to meet. Humans do not create from nothing; they must employ the principles of engineering and physics and their experience. Fromson v. Advance Offset Plate, Inc., 755 F.2d at 1556 n. 3, 225 U.S.P.Q. (BNA) at 31 n. 3. It cannot be the law that the only inventions patentable are those that cannot be explained by any known principles of engineering or physics. This court understated the point in Lindemann Maschinenfabrick GMBH v. American Hoist & Derrick Co., 730 F.2d at 1462, 221 USPQ at 489, when it said: "That the claimed invention may employ known principles does not in itself establish that the invention would have been obvious. Most inventions do."

Fourth, the "principles" and "experience" here referred to were those taught the court by the inventor Caveney. There was no evidence on which it could be said that one deprived of Caveney's testimony would have been led by general principles and common experience to make the claimed inventions in suit. Indeed, the evidence was to the contrary. The district court's reliance on general principles and common experience simply ignored what actually happened in the cable tie industry over many years, including Dennison's failed decade of research. It also ignored the testimony of Dennison's engineers Stewart and Joyce, supra, in which they described the difficult nature of cable tie design work.

Whatever the unidentified "principles" of engineering and physics the court intended, and whatever was the non-described "common experience of mankind" the court had in mind, those principles and that experience were fully available to all who tried and failed before Caveney succeeded. They cannot serve as a basis for holding that the claimed inventions in suit would have been obvious under Sec. 103.

(b) Unexpected Results

During closing argument, after the 11-day recess, Dennison's counsel handed the trial court a copy of this court's decision and opinion in In re Sernaker, 702 F.2d 989, 217 U.S.P.Q. (BNA) 1 (Fed.Cir.1983) and implied that this court required an "unexpected result" in support of nonobviousness, saying that Sernaker "capsulizes the feelings of the Court of Appeals for the Federal Circuit." Dennison's counsel then went on to cite Sernaker as requiring an unexpected result in relation to each patent in suit, saying, for example, in respect of the '538 patent:

... it is exactly what In re Sernaker says is not an invention [sic, is not patentable] because what you achieve by adding a hinge is exactly what you expect to achieve. You achieve flexibility. You do not get an unexpected result.

As above indicated, the district court, before the recess, noted that "other people were going in different ways" and "did not arrive at the Caveney structure." The court also said at that time that he was impressed that "other people faced with the same problem did not arrive" at Caveney's result, and that the strongest argument against randomly combining features of the prior art was "what other people were doing contemporaneously." After hearing the argument of Dennison's counsel based on Sernaker, the district court proceeded to find that Caveney's inventions produced no "unexpected result".

Dennison's counsel misstated the law to the district court. Sernaker does not hold that an "unexpected result" is required before a conclusion of nonobviousness may be reached, and does not "capsulize" this court's "feeling" that such a requirement exists. On the contrary, this court has specifically pointed out that, while an "unexpected result", like "synergism", may be evidence of nonobviousness, it is not a requirement. See Lindemann Maschinenfabrick GMBH v. American Hoist & Derrick Co., 730 F.2d at 1461, 221 USPQ at 488; American Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1360-61, 220 U.S.P.Q. (BNA) 763, 771 (Fed.Cir.1984).23

Moreover, if there were a requirement for an "unexpected result", that requirement would have been fully met here. That the achievement of a 1/2 to 80 pound ratio between insertion and withdrawal forces was unexpected is fully reflected in the totality of the evidence. That that achievement was not mentioned by the district court reflects not its absence but the court's overconcentration on a superficial review of the prior art and disregard of the claimed structural characteristics and relationships that produced that achievement.

(6) Objective Evidence

The human, real world story in evidence here not only reflects the inadequacy of the prior art, but compels a conclusion of nonobviousness of the claimed inventions in suit.

As above indicated, the cable tie of the patents in suit was an outstanding commercial success, selling at over $50 million a year by 1984 and becoming, early on, the unquestioned leading tie on the market. With all that waiting on the successful invention of a superior cable tie, the district court did not explain why, if Caveney's superior cable tie would have been so obvious, the many inventors that preceded Caveney did not find it so.

It is extremely difficult to believe that Panduit and Dennison would have invested years and millions of dollars in an effort to achieve what had been obvious all along to all of ordinary skill in the art.

That many others, including Dennison, had tried for years and failed to create a superior cable tie is virtually irrefutable evidence that the superior tie of the patents in suit would not have been obvious to those skilled in the art when it was invented.

That earlier workers were, as the district court correctly said, "going in different ways" is strong evidence that inventor Caveney's way would not have been obvious.

That Dennison, a large corporation with many engineers on its staff, did not copy any prior art device, but found it necessary to copy the cable tie of the claims in suit, is equally strong evidence of nonobviousness. As was once so well said, of an infringer and the invention appropriated: "It gives the tribute of its praise to the prior art; it gives the [invention] the tribute of its imitation, as others have done." Diamond Rubber Co. v. Consolidated Rubber Tire Co., 220 U.S. 428, 441, 31 S. Ct. 444, 450, 55 L. Ed. 527 (1911).

It is at least incongruous for a court to say a commercially successful, industry-leading invention would have been obvious (to it) in view of the prior art and "principles/experience", when the record clearly establishes that when the invention was made it had not for years been obvious to earlier workers in the art, to the defendant's experts, or to the Examiner--all of whom had before them the same prior art and the same "principles/experience".

Dennison's brief limits its argument on the objective evidence essentially to an attack on the commercial success of the cable tie of the patents in suit.24 That attack is without merit. First, it comes with poor grace from an infringer who achieved its own great commercial success and became the industry's second or third leader by making and selling its copy of that cable tie. Second, Dennison's suggestion that Panduit's success is due to unclaimed features or to its advertising or sales staff are either unsupported or contradicted in the evidence. Third, Dennison's attempt to limit Panduit's success to the tie claimed in the '538 patent makes no sense. That tie, and Dennison's copy of it, embodies all inventions set forth in all of the asserted claims in suit, and all are entitled to the benefit of the overwhelming success to which each has contributed.

That respect is required for the effect to be given objective evidence of nonobviousness, when that evidence is present, has been stated by this court. W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1555, 220 U.S.P.Q. (BNA) 303, 314; Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1538, 218 U.S.P.Q. (BNA) 871, 879 (Fed.Cir.1983).

The objective evidence is here so strong as to have withstood prior art references substantially more pertinent than those relied upon by Dennison. If it was ultimately given any weight by the district court, that weight was insufficient. In evaluating the objective evidence and its thrust, the district court erred.

(c) Conclusion On Sec. 103

The district court having erred in reaching its conclusion on obviousness under Sec. 103, the portion of its judgment resting on that conclusion must be reversed.

(II) Double Patenting

The district court, as above indicated, devoted two sentences to Dennison's double-patenting defense. It did not, nor does Dennison, compare the claims, or assert that the claims of the '869 and '538 patents are not patentably distinct, or that issuance of the '538 patent extended the term of the '869 patent. See In re Longi, 759 F.2d 887, 892-95, 225 U.S.P.Q. (BNA) 645, 648-50 (Fed.Cir.1985).

Nothing of record supports a double patenting defense. The court in effect simply and erroneously held the '538 patent invalid for obviousness under 35 U.S.C. Sec. 103 in view of the '869 patent, and under the rubric of "double patenting". Because the '869 patent is not prior art to the '538 patent, supra, n. 11, the patented invention cannot be held to have been obvious under Sec. 103. Hence, that portion of the district court's judgment that rests on its double patenting determination must be reversed.25 Moreover, had the district court properly applied the test for obviousness type double patenting by comparing the claims of both patents, it would have concluded that the '538 claimed invention is not an obvious variation of the '869 claimed invention.

(III) The '146 Patent and Sec. 102(g)

Dennison, as it has a right to do, attempts to support the judgment with respect to the '146 patent by alleging error in the district court's rejection of its Sec. 102(g) defense against that patent. The trial court found that Dennison was in no way prejudiced by a delay in the filing of the '146 patent application. Nothing of record would indicate that that finding was clearly erroneous.

Dennison's effort to create a forfeiture out of the instant facts, arguing that Panduit "lost the patent right" by delay is meritless, as indicated by Paulik v. Rizkalla, 760 F.2d 1270, 1272, 226 U.S.P.Q. (BNA) 224, 225 (Fed.Cir.1985) (in banc), wherein it was stated:

This historic jurisprudence from which 35 U.S.C. Sec. 102(g) flowed reminds us that "the mere lapse of time" will not prevent the inventor from receiving a patent. Mason v. Hepburn, 13 App.D.C. 86, 91, 1898 C.D. 510, 513 (1898). The sole exception to this principle resides in section 102(g) and the exigencies of the priority contest.

There is thus no basis for applying section 102(g) to mere delays in filing, except in priority contests where the equities between claimants for the same invention may be fully evaluated.

Dennison tries to extend 102(g) to the present facts, arguing that Bourne invented the "concept" of the '146 patent. However, as above indicated, n. 8, supra, Bourne discloses a non-hinged pawl. "Concepts" do not anticipate. The law looks to the particular inquiries set forth in 35 U.S.C. Sec. 102(a)-(g), which focus on knowledge, use, sale, disclosure, etc., of the invention. Notions of "concept", "essence", "key", "gist", etc., are no more useful in the context of Sec. 102 than elsewhere, because they divert the fact-finder's attention from the subject matter of the invention as a whole.

"Anticipation" for the purposes of Sec. 102 requires "the presence in a single prior art disclosure of all elements of a claimed invention arranged as in that claim". Connell v. Sears, Roebuck & Co., 722 F.2d 1542, 1548, 220 U.S.P.Q. (BNA) 193, 198 (Fed.Cir.1983). It is a finding of fact subject to the clearly erroneous rule. Carman Industries, Inc. v. Wahl, 724 F.2d 932 (Fed.Cir.1983); W.L. Gore & Assocs., Inc. v. Garlock, Inc., 721 F.2d at 1548, 220 USPQ at 309. The district court's refusal to find the '146 patent anticipated by Bourne was not clearly erroneous.

Where, as here, an alleged other inventor discloses at most a contemporaneous development distinct from the invention in suit, there is no basis for Dennison's suggestion that an extension of section 102(g) be created for application outside a priority contest.

(IV) Dennison's Presentation on Appeal

On appeal, Dennison proffers only superficial characterizations of the claimed inventions and the references, as it did before the district court. At no point does Dennison treat any claimed invention or any reference as a whole. The meaningless one-word descriptions of the claimed inventions and reference structures in Dennison's brief are of no use or assistance to this court.26

Dennison misstates the record in arguing that the prosecution history of the '538 patent indicates that the hinge there shown and claimed will "collapse under heavy load." The quoted phrase related to an entirely different embodiment that was ruled new matter and never added to the '538 application.

The trial court relied primarily on the '869 patent as prior art against the '538 patent. It twice specifically and correctly refused to rely on the Orban reference. Dennison's brief relies, ineffectively and without reference to the district court's view, on elements it purports to find in Orban.

In a confused and confusing effort to shore up the obviousness conclusion of the district court, Dennison abandons much of what the court said. It then cites items in the record about which not a word was said at trial, and with no explanation as to how those items can be seen to suggest the claimed inventions in suit. See Amstar Corp. v. Envirotech Corp., 730 F.2d 1476, 1486, 221 U.S.P.Q. (BNA) 649, 656-57 (Fed.Cir.1984).

Dennison's analysis here of the prior art relied upon by the district court as invalidating the '869 patent consists of two lines in which it says only that three prior patents "teach support." That useless characterization does not aid this court in analyzing the references as wholes, nor does it begin to explain how the cited references teach or suggest the claimed inventions. Similarly, Dennison points to an incomprehensible phrase ("flexibility pivotable ratchet buckle"), in the Fein reference, while ignoring Fein's full disclosure and the trial record, which establish that Fein's pawl is rigid and that its structure is entirely distinct from those claimed in the patents in suit.

Dennison intimates here that a cable tie found on sale was that of the Bourne patent, but Dennison's counsel admitted to the district court that "pictures in the on sale defense are of different cable ties." The district court stated that Dennison made no serious argument before it that the on sale product anticipated the claimed inventions in suit.

Dennison misleadingly says that the inventions of the '869 and '538 patents were made "simultaneously", referring to a stipulation that certain drawings established conception of the subject matter of the representative claims "as of" January 1968, which is an entirely different matter. Dennison knows the inventions were not made simultaneously, admitting at another point in its brief that the first sketch of the '869 structure was made a few weeks after the '538 structure was reduced to practice.

In a post-hearing letter, Dennison's counsel tells the court that an exhibit (PX-55) was "excluded from evidence at the trial". That is not true. The deposition testimony forming part of that exhibit, and the documentary exhibits referred to therein, were admitted into evidence. There was objection to the argument portion of the exhibit. The district court accepted that portion as a trial brief and invited Dennison to respond. Dennison did not do so.

In sum, Dennison's presentation on appeal is disingenuous, containing mischaracterizations, misleading statements, and improper submissions. It has unnecessarily burdened the court with extraordinary need to check the record in respect of each of its assertions, only to find in too many instances a lack of candor. Accordingly, Panduit is awarded double its costs on appeal. Amstar Corp. v. Envirotech Corp., 730 F.2d at 1486, 221 USPQ at 656-57.

The judgment of the district court with respect to validity is reversed and the case is remanded for further proceedings not inconsistent with this opinion.

REVERSED.

24. An integral one-piece cable tie to be tensioned about a bundle of wires and the like, said cable tie comprising an elongated flexible strap, a frame integral with one end of said strap and including an abutment wall, said frame having an entry surface and an exit surface and a strap-receiving opening extending therethrough, a row of teeth disposed on one longitudinal surface of said strap and arranged transversely with respect thereto, a pawl hingedly mounted on and integral with said frame and extending into said strap-receiving opening toward said abutment wall, said abutment wall having a strap-bearing surface disposed toward said pawl and defining therewith a strap-receiving channel, and a set of teeth on said pawl arranged transversely with respect thereto and disposed toward said abutment wall and shaped complementary to said row of teeth on said strap, the crests of a plurality of the teeth in said set of teeth lying in a common surface space equidistantly from said strap-bearing surface in the tensioned condition, said strap being deformable into a loop encircling a bundle of wires with the other end of said strap extending into said strap-receiving channel and through the opening in said frame and therebeyond, said set of teeth being disposed toward said row of teeth as said strap is tightened about the bundle of wires to a tensioned condition, said set of teeth upon release of said tensioned strap moving into firm engagement with adjacent ones of said row of teeth, any force tending to withdraw said strap from within said strap-receiving channel in a strap-loosening direction serving to move said set of teeth including said plurality of teeth into more firm engagement with the engaged ones of said row of teeth firmly to wedge said strap between said strap-bearing surface and said pawl, whereby to prevent inadvertent withdrawal of said strap from said frame and thus to lock said strap in its tensioned condition about the bundel [sic] of wires.

10. An integral one-piece cable tie to be tensioned about a bundle of wires and the like, said cable tie comprising an elongated flexible strap, a row of abutments disposed on one longitudinal surface of said strap and arranged transversely with respect thereto, a frame integral with one end of said strap and including an end wall and an abutment wall and a pair of longitudinally extending and spaced-apart side walls interconnecting said end wall and said abutment wall, said frame having an entry surface and an exit and a strap-receiving opening extending therethrough, a ledge on said end wall extending between and interconnecting said side walls and extending longitudinally from said end wall toward said abutment wall, a pawl disposed within said frame in said strap-receiving opening and pivotally mounted on and integral with said ledge, said abutment wall having a strap-bearing surface disposed toward said pawl and defining therewith a strap-receiving throat, the longitudinal axis of said throat being disposed substantially normal to the longitudinal axis of said strap as molded, and a tooth on said pawl arranged transversely with respect thereto and disposed toward said abutment wall and shaped to engage said row of abutments on said strap, said strap being deformable into a loop encircling a bundle of wires with the free end of said strap extending into said strap-receiving throat and through the opening in said frame and therebeyond, said tooth being disposed toward said row of abutments as said strap is tensioned about and release [sic] said strap causing at least one of said abutments firmly to engage said tooth, any force tending to withdraw said strap from within said strap-receiving throat in a strap-loosening direction serving to move said tooth into more firm engagement with the engaged ones of said row of abutments firmly to grip said strap between said strap-bearing surface and said pawl, whereby to prevent inadvertent withdrawal of said strap from said frame and thus to lock said strap in its tensioned condition about the bundle of wires.

(c) The '538 Patent

Claims 1 and 6 were stipulated as representative:

1. An integral one-piece cable tie to be tensioned about a bundle of wires and the like, said cable tie comprising an elongated flexible strap, a row of abutments disposed on one longitudinal surface of said strap and arranged transversely with respect thereto, a frame integral with one end of said strap and including an end wall and an abutment wall, said frame having an entry surface and an exit surface and a strap-receiving opening extending therethrough, a ledge on said end wall extending longitudinally therefrom toward said abutment wall, a pawl disposed within said frame in said strap-receiving opening between said ledge and said exit surface, a hinge interconnecting said ledge and the end of said pawl disposed toward said entry surface, the longitudinal extent of said hinge being less than the longitudinal extent of said ledge at the junction therebetween, the longitudinal cross-sectional area of said hinge being less than the longitudinal cross-sectional area of said pawl at the junction therebetween, said pawl and said hinge cooperating so that said pawl is limited to substantially pivotal movement with respect to said frame about the connection to said ledge, said pawl in the as-molded position thereof and in all other positions thereof including the tensioned position thereof being disposed entirely within said frame between the entry and exit surfaces thereof, said abutment wall having a strap-bearing surface disposed toward said pawl and defining therewith a strap-receiving throat, and a tooth on said pawl arranged transversely with respect thereto and disposed toward said abutment wall and shaped to engage said row of abutments on said strap, said strap being deformable into a loop encircling a bundle of wires with the free end of said strap extending into said strap-receiving throat and through the opening in said frame and therebeyond, said tooth being disposed toward said row of abutments as said strap is tensioned about the bundle of wires to a tensioned condition, and release of said strap causing at least one of said abutments firmly to engage said tooth, any force tending to withdraw said strap from within said strap-receiving throat in a strap-loosening direction serving to move said tooth into more firm engagement with the engaged ones of said row of abutments firmly to grasp said strap between said strap-bearing surface and said pawl, whereby to prevent inadvertent withdrawal of said strap from said frame and thus to lock said strap in its tensioned condition about the bundle of wires.

6. The integral one-piece cable tie set forth in claim 1, wherein the normal tensioned position of said pawl is substantially the as-molded position thereof.