concurring.
I fully concur in the judgment of reversal and with the opinion of the Court but want to express some additional views because of the concurring opinion of my Brother Brennan and the dissenting opinion of my Brother Harlan. The latter two opinions, in my judgment, attempt to introduce wholly unnecessary and undesirable confusions, intricacies and complexities into what has been essentially a very simple question under the opinions of this Court, and that is: How can a court decide whether a person who has bought and owns a patented commodity composed of a combination of unpatented *347elements is actually making a new one so as to infringe the patent, rather than merely replacing a worn-out part or parts necessary to continue the use of the commodity, which does not constitute patent infringement? I put the question as one of direct infringement because I agree with the other three opinions in this case that the petitioner Aro Manufacturing Co. here who cut out and sold fabrics to replace worn-out covers in patented automobile tops can be guilty of contributory infringement only if automobile owners who buy the fabrics and use them in their own cars are themselves guilty of direct infringements and liable for treble damages.1 The crucial question here therefore is one of direct, not contributory, infringement. For this reason it seems to me that most of the talk in the case about contributory infringement and misuse of patents is confusing and beside the point. For the same reason the emphasis in this Court, the District Court and the Court of Appeals on the recodification 2 of the patent laws in 1952 seems to me to be out of place. The language and history of that Act show plainly: *348(1) that Congress wanted to continue in force, but not expand, the judge-made doctrine of contributory infringement under which a person who knowingly aids, encourages or abets the direct infringement of a patent is to be held liable as a contributory infringer; 3 (2) that Congress *349did not want patentees to be barred from prosecuting their claims for direct infringement merely because they exercised their right to assert a claim in or out of court for contributory infringement;4 (3) that the long-existing scope of a patentee’s monopoly rights was not to be expanded *350beyond what it had always been, that is, the exclusive right to make, use or sell a patented invention during the life of the patent.5 The present case, therefore, narrows down to this and no more: Is the making or the use of a new piece of fabric to replace the worn-out fabric cover of an automobile top attached to an automobile that the owner has bought and paid for as his own an infringement of a patent on that top? Note immediately that if it is an infringement it must be because the owner of the top, in doing nothing but replacing the worn-out fabric, “makes” the patented invention which is the top with all its parts. Common sense and prior decisions of this Court require us to hold that an automobile owner does not “make” a whole top when he merely replaces its worn-out fabric cover.
Let us first take a quick look at the top described in the patent. It is composed first and foremost of a metal frame, no part of which is, or could have been, patented. It also includes several other pieces of metal, one of them called a “wiper,” which is simply a plain piece of metal that is not and could not have been patented. Again, *351there is a piece of fabric cut in a certain shape to fit and cover the unpatented frame. Of course this fabric could not be patented unless it had some peculiar novel quality of its own, but no such characteristic is or could have been claimed. The District Court held that this aggregation of nonpatentable parts was patentable as achieving a new result. The Court of Appeals, without passing on this question at all, merely took it for granted that a patentable “discovery” had been made. I shall also act on that assumption although I am not sure in just what respect the aggregation of these common components could possibly have served a new purpose or have been the result of anything more than the simplest childlike mechanical skill. In fact, the patentee must have known all about the old-fashioned surrey with the fringe on top and with isinglass curtains you could roll right down in case of a change in the weather. The top is also reminiscent of the tops of Model T Fords which began to scare horses on country roads nearly half a century ago. A reading of the record indicates that the new discovery might be thought to inhere in the fact that (a) this top protects from rain better than any that had previously been manufactured, (b) it could be made rainproof from the inside rather than requiring some one to get outside and manipulate snaps, or (c) the folding material is fastened “below the top of the tonneau” and “a wiping arm automatically operated by the bow structure [is used] for wiping the inside of the folding material as the top is raised and pressing it against the top of the tonneau.” None of these alleged additions to the present stock of surrey and Tin Lizzie knowledge, however, gives us much aid in determining the real issue being decided here, which is whether replacing the fabric amounts to the “making” of a new top and thus a direct infringement of the patent. It is of importance in considering this question that the District Court found, and *352the Court of Appeals agreed, that the only thing the defendant made was the fabric portions of convertible tops. Certainly I suppose it would not be claimed that the sale of unshaped woolen fabric — a staple article of commerce — would make a tailor or merchant guilty of contributory infringement if he sold it to the owner of a pair of patented trousers for patching purposes. But evidently the contention is that when petitioner here put his scissors or whatever cutting instrument he used on the fabric and shaped it so that it would fit the top, he was thereby aiding the top’s owner to “make” a new top, or at least to “reconstruct” one from the ground up. Nothing in any of the prior cases of this Court indicates that such a contention should be sustained.
The case that may well be classified as the leading one on the subject of “making” or “reconstruction” as distinguished from “repair” is Wilson v. Simpson, 9 How. 109, decided in 1850. That case involved a patent on a planing machine composed of unpatented parts described as the frame, the cogwheels, the shaft, and other elementary parts, which, when put together, constituted what was known as the “Woodworth planing-machine.” Able counsel argued for the patentee that the cutting knives, which had to be replaced from time to time, could not be replaced without thereby infringing the patent. The contention was that the machine ceased to exist or have any “material existence” the moment its knives wore out, and that for this reason replacement of the knives amounted to a “making” of the whole patented invention which infringed the patentee’s exclusive right to “make, use, and vend.” The Court refused to accept this concep-tualistic and misleading argument as to when a tangible machine ceases to. have a “material existence.” It did agree that “when the material of the combination ceases to exist, in whatever way that may occur, the right to *353renew it depends upon the right to make the invention. If the right to make does not exist, there is no right to rebuild the combination.” Id., at 123. But the Court then went on to enunciate what has been the accepted rule in this Court ever since with respect to component parts of the combination: “When the wearing or injury is partial, then repair is restoration, and not reconstruction. . . . And it is no more than that, though it shall be a replacement of an essential part of a combination.” Ibid.
In further explanation the Court pointed out that “[f]orm may be given to a piece of any material, — wood, metal, or glass, — so as to produce an original result, or to aid the efficiency of one already known, and that would be the subject for a patent.” It went on to say that if that patented article should happen to be broken so that its parts could not be readjusted or so worn out and beyond repair as to be wholly useless, then a purchaser could not make another to replace it without infringing, but would have to buy a new one. This because the Court said that would amount to an “entire reconstruction” of the patented article. “If,, however,” and this is of crucial importance with reference to the combination patent in the present case, “this same thing is a part of an original combination, essential to its use, then the right to repair and replace recurs.” Id., at 124.
The common-sense rule of Wilson v. Simpson was followed in Heyer v. Duplicator Mfg. Co., 263 U. S. 100. Involved there was a combination patent, a multiple copying machine, “one element of which,” the Court said, “is a band of gelatine to which is transferred the print to be multiplied and which yields copies up to about a hundred. This band is attached to a spool or spindle which fits into the machine. Anyone may make and sell the gelatine composition but the ground of recovery was that the de*354fendant made and sold bands of sizes fitted for use in the plaintiffs machine and attached them to spindles, with intent that they should be so used. The main question is whether purchasers of these machines have a right to replace the gelatine bands from any source that they choose. If they have that right the defendant in selling to them does no wrong.” Id,., at 101. A unanimous Court, speaking through Mr. Justice Holmes, recognized that the rule of Wilson v. Simpson disposed of this question, saying that “[s]ince Wilson v. Simpson, 9 How. 109, 123, it has been the established law that a patentee has not ‘a more equitable right to force the disuse of the machine entirely, on account of the inoperativeness of a part of it, than the purchaser has to repair, who has, in the whole of it, a right of use.’ ” 263 U. S., at 101. Like the bands of gelatine in Heyer, the fabric in the present case is a common article of commerce, and also like the gelatine it has to have a special form to fit into the combination.
None of this Court’s cases relied upon by my Brothers Harlan and Brennan justifies adoption of the supposed guides and standards referred to in their opinions. Deciding whether a patented article is “made” does not depend on whether an unpatented element of it is perishable, or how long some of the elements last, or what the pat-entee’s or a purchaser’s intentions were about them, regardless of whether the application of such standards is considered a question of law, as it is by Mr. Justice Brennan, or a question of fact, as it is by Mr. Justice Harlan. The holdings of prior cases in this Court have not actually rested on such minor and insignificant factors in determining when an article has been “made” and when it has not. A case in which only a minimal component has been omitted from a “making” of the combination, such as the extreme example of omission of a single bolt from a patented machine or a button from a patented garment, might call upon this Court to articulate some rather ob*355vious refinements to this simple test of “making.” 6 But this is by no means such a case and it is dangerously misleading to suggest that so clear a case as this one requires the application of a Pandora's flock of insignificant standards, especially when it is recognized, as it must be upon analysis, that consistent application of the standards suggested would actually change the basic test from “making” to something not satisfactorily defined but indisputably different. And surely the scope of a patent should never depend upon a psychoanalysis of the patentee’s or purchaser’s intentions, a test which can only confound confusion.7 The common sense of the whole matter is, as recognized in the Wilson case and again in the opinion of the Court today, that in none but the most extraordinary case — difficult even to imagine — will a court ever have to invoke specially contrived evidentiary standards to determine whether there has actually been a new “making” of the patented article.
Neither Leeds & Catlin Co. v. Victor Talking Machine Co., 213 U. S. 325, nor Cotton-Tie Co. v. Simmons, 106 U. S. 89, nor any other cases of this Court that have treated this subject, depart from this common-sense rule of Wilson v. Simpson. In fact, as Mr. Justice Holmes pointed out in Heyer v. Duplicator Mfg. Co., supra, the question in Leeds & Catlin did not concern “a right to sub*356stitute worn out parts” and “[t]he authority of Wilson v. Simpson and the cases that have followed it was fully recognized [in Leeds & Catlin] and must be recognized here.” 263 U. S., at 102. The Cotton-Tie case likewise contains no support for holding that replacing the fabric in this top amounts to a “making” or even a “remaking” of the whole convertible top. The patent involved there was a cotton bale tie. Marked on each, for whatever it was worth, was “Licensed to use once only.” After these ties had been used once and broken off the cotton bales, the old tie material, along with the buckles, was sold as scrap, then re-rolled, straightened, riveted together, and cut into proper lengths and attached to an old buckle. It was this making of completely new ties out of the discarded scrap material that was held to amount to a making of the patented device and therefore an infringement.8 Cf. Morgan Envelope Co. v. Albany Paper Co., 152 U. S. 425, 433-434. Morgan Envelope, in addition to its distinction of the Cotton-Tie case, is notable for its explicit reaffirmance of Wilson v. Simpson and for its statement by a unanimous Court that “[t]he true distinction” is whether the component part of the combination is “the subject of a separate patent.” 152 U. S., at 435. I agree with my Brother Whittaker that this remains the true distinction today.9
*357In my judgment it would create mischievous results for a majority of this Court to create or approve ambiguous evidentiary standards which could only obfuscate the simple fact of whether a person is “making” a patented article composed of old unpatentable elements. For example, there should be no attempt to decide whether there is a making by comparing the time that the different elements of such a patent normally will exist if let alone. The owner is under no obligation to let them alone. Every owner has the right to repair and patch each part of the property he bought and paid for in order to make it last as long as he can. Everyone knows that this patented top is likely to last as long as the car itself if it is repaired from time to time. An accident might wholly destroy the entire top and then it might have to be replaced by a new one, which if done without the patentee's consent would of course be an infringement. I cannot suppose, however, that if the bows upon which the fabric rests at the top, or the metal frames upon which it stands, or the metal wiper which helps to level off the curtain folds, or the snaps that may have to be used, should happen to become worn out or destroyed, that anyone could reasonably come to the conclusion that a replacement of one of these worn-out parts would be a complete rebuilding of the old top or the “making” of a new one and therefore an infringement of the patent. Such a contention would seem little short of fantastic to me, and the same is true as to replacement of the fabric.
This case is of great importance in our competitive economy. The record .shows that petitioner is but one of many small business enterprises filling a useful place in manufacturing the comparatively smaller parts of larger products like automobiles. It is quite right and in keeping with our patent system that small business enterprises should no more than large enterprises be allowed to *358infringe the patents of others. But businessmen are certainly entitled to know when they are committing an infringement. It is for that reason that the patent statutes require applicants to define with particularity and claim without ambiguity the subject matter which is regarded to be an invention. 35 U. S. C. §§ 112, 154. In the absence of suits like this how could á businessman even suspect that he might be infringing a patent by marketing an unpatented and unpatentable part of a combination like this fabric top? It has long been settled with respect to combination patents that the monopoly rights extend only to the patented combination as a whole and that the public is free to appropriate any unpatented part of it, “however important.” Special Equipment Co. v. Coe, 324 U. S. 370, 376, and cases cited. And in the 1952 recodification Congress was very careful, in what is now 35 U. S. C. § 271 (c), to specify that a vendor is liable as a contributory infringer only if he has sold the component part “knowing” that it is to be used in an infringement.10 But to what avail these congressional precautions if this Court, by its opinions, would subject small businessmen to the devastating uncertainties of nebulous and permissive standards of infringement under which courts could impose treble damages upon them for making parts, distinct, separable, minor parts, or even major parts of a combination patent, upon which parts no patent has been or legally could have been issued. The fact that subjection of a small business enterprise to treble damages for such activities can have disastrous or even lethal consequences is suggested by the observation of the *359District Court in this very case that petitioner had not made a supersedeas bond, whether because unable or unwilling to do so the court said it was not informed. The efforts of Congress to help small business enterprises keep their heads above water in the Small Business Act of 1953,11 which was enacted almost contemporaneously with the recodification of the patent laws, could be frustrated at least in large part by subjecting those engaged in industrial activities to the constantly overhanging threat of suits for patent infringements for the sale of unpatented articles. And I think it is indisputable that the unpredictability in any given case of the resolution of the intricate standards suggested by my Brothers Harlan and Brennan would deter small businessmen from engaging even in activities that all members of this Court would ultimately agree do not constitute contributory infringement. In our own case these standards led to opposite conclusions when applied by three distinguished judges of the Court of Appeals and when applied by Mr. Justice Brennan.
The established policy in this Nation for more than a century has been that when an article described in a patent is sold and “passes to the hands of the purchaser, it is no longer within the limits of the monopoly. It passes outside of it, and is no longer under the protection of the act of congress. . . . Contracts in relation to it are regulated by the laws of the State, and are subject to state jurisdiction.” 12 In this day of advanced technology and mechanical appliances upon which so many people depend, *360this wise policy against permitting patentees to expand their control of commodities after they reach the hands of bona fide purchasers is all the more important. Congress surely did not intend for it to be left within the sole discretion of the patent monopolist whether an unpatented component part will be separately available to the purchaser for replacement in the combination or whether, when that part wears out, the purchaser will be forced to replace a larger subcombination of the patented product or perhaps even the entire aggregation. I have an idea it would greatly surprise and shock the owners of convertible automobiles throughout this country to learn that there is a serious contention made that if they buy a new fabric to replace the worn-out fabric on their convertible top, loose legal formulas like those suggested by my Brothers Harlan and Brennan can be applied to make them liable for treble damages, attorneys’ fees and heavy costs. No such doctrine should be allowed to gain a foothold through judicial decisions. If bona fide purchasers of goods throughout the Nation are to be subjected to any such phenomenal expansion of the right of patentees, Congress, not this Court, should do it. One royalty to one patentee for one sale is enough under our patent law as written. And this proposition can be stated with knowledge that there is not a single sentence, clause, phrase or word in the entire legislative history of the 1952 recodification which could have led Congress to believe that it was being asked by means of the provisions enacted to enlarge the scope of a patentee’s exclusive right to “make, use, and vend” the thing patented.
A fundamental error underlying the misleading standards suggested by my Brothers Harlan and Brennan is the notion that in a case of this kind a court is obliged to search for the alleged “heart” or “core” of the combination patent. This misconception, which has unequivo*361cally been rejected in our prior decisions,13 is nothing but an unwarranted transformation and expansion of a combination patent, which was correctly described by Mr. Justice Jackson as conferring only a'“. . . right in an abstruse relationship between things in which individually there is no right.” 14 A patented combination is no more than that, a novel relationship brought to bear on what presumably are familiar elements already in the public domain. Such familiar elements are not removed from the public domain merely because of their use, however crucial, in the novel combination. Of course, if novelty should inhere in one of the parts as well as in the whole, then that novel “heart” or “core” can be separately patented and separately protected. But in the absence of such a separate patent, which in this case would presumably be some sort of patent on the utility of the shape of the fabric, the public has the right to make, use and vend each part subject only to the established limitation of contributory infringement: that a part may not be knowingly supplied for use by an unauthorized person in a new making of what is in effect the whole combination.
Finally, petitioners point out that there is a serious constitutional question involved in the claim that automobile owners can be mulcted for damages for replacing the worn-out fabric in their convertible tops. Article I, § 8 of the Constitution provides:
“The Congress shall have Power ... To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.”
*362I have no doubt of the wide powers of the States to govern the sale of fabrics, clipped or unclipped, cut or uncut, cloth, oilcloth, isinglass or plastic. But it is difficult for me to think that shaping up a common piece of fabric with a common pair of scissors or other cutter can be exalted to that important category of “Discoveries” that the Constitution authorizes Congress to promote by special federal legislation. Nor do I believe that a purpose should be attributed to Congress to allow the courts, by means of an incalculable weighing of a complex of nebulous standards, to divine whether persons should be given monopoly rights on account of the way they trim ordinary fabrics. If there is anything sure about the 1952 recodification it is that the purpose was to clarify not to confuse the law. The test applied in the opinion of the Court today will not confuse, is in accord with our prior decisions endorsed by Congress and, most important, is in keeping with the general purposes of this Nation to retain an economy in which competition is a general law of trade except where actual “Discoveries” have been made.
This elementary principle was brought to the attention of Congress by Mr. Giles S. Rich, the chief draftsman of the provisions on contributory infringement in the 1952 recodification:
“Mr. Rich. ... I should state at the outset that wherever there is contributory infringement there is somewhere something called direct infringement, and to that direct infringement someone has contributed. It is a very different thing from a concept like contributory negligence.” Hearings before Subcommittee of House Judiciary Committee on H. R. 3760, 82d Cong., 1st Sess. 151 (1951).
If anyone is inclined despite other evidence to the contrary to attribute to Congress a purpose to accomplish any far-reaching changes in the substantive law by this enactment, he should take note that just before the bill was passed in the Senate, Senator Salton-stall asked on the floor, “Does the bill change the law in any way or only codify the present patent laws ?” Senator McCarran, Chairman of the Judiciary Committee which had been in charge of the bill for the Senate, replied, “It codifies the present patent laws.” 98 Cong. Rec. 9323 (July 4, 1952).
To this end, the 1952 Act provides (35 U. S. C. § 271):
“(b) Whoever actively induces infringement of a patent shall be liable as an infringer.
“(c) Whoever 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 non-infringing use, shall be liable as a contributory infringer.”
The hearings and committee reports show that the only purpose of these provisions was to prevent this Court’s decision in Mercoid Corp. v. Mid-Continent Inv. Co., 320 U. S. 661, from being treated as having obliterated the law of contributory infringement:
“Mr. Rogers. Then I take it from your statements, that there is some difference of opinion among those engaged in the practice of patent law as to whether or not the Supreme Court in its decisions has done away with contributory infringement.
“Mr. Rich. There is a great difference of opinion on the part of the bar and also apparently on the part of the judiciary ....
“Mr. Rogers. Then in effect this reeodification, particularly as to section 231 [from which came the present §271], would point out to the court, at least that it was the sense of Congress that we remove this question of confusion as to whether contributory infringement existed at all, and state in positive law that there is such a thing as contributory infringement, or at least it be the sense of Congress by the enactment of this law that if you have in the Mercoid case done away with contributor}'- infringement, then we reinstate it as a matter of substantive law of the United States and that you shall hereafter in a proper case recognize or hold liable one who has contributed to the infringement of a patent.
“That is the substantive law that we would write if we adopted this section 231 as it now exists. Is that not about right?
“Mr. Rich. That is a very excellent statement . . . Hearings, supra, note 1, at 159. [Footnote S continued on p. S49-]
*349"... Considerable doubt and confusion as to the scope of contributory infringement has resulted from a number of decisions of the courts in recent years. The purpose of this section is to codify in statutory form principles of contributory infringement and at the same time eliminate this doubt and confusion.” H. R. Rep. No. 1923 on H. R. 7794, 82d Cong., 2d Sess. 9.
To protect a patentee’s right to sue for contributory infringement, the 1952 Act provides:
“(d) No patent owner otherwise entitled to relief for infringement or contributory infringement of a patent shall be denied relief or deemed guilty of misuse or illegal extension of the patent right by reason of his having done one or more of the following: (1) derived revenue from acts which if performed by another without his consent would constitute contributory infringement of the patent; (2) licensed or authorized another to perform acts which if performed without his consent would constitute contributory infringement of the patent; (3) sought to enforce his patent rights against infringement or contributory infringement.” 35 U. S. C. § 271 (d).
This provision was designed specifically to prevent the Mercoid case from being interpreted to mean that any effort to enforce a patent against a contributory infringer in itself constitutes a forfeiture of patent rights:
“Mr. Rich. . . .
“Other decisions following Mercoid have made it quite clear that at least some courts are going to say that any effort whatever to enforce a patent against a contributory infringer is in itself misuse. The cases are cited in the old hearings. Therefore, we have always felt— we who study this subject particularly — that to put any measure of contributory infringement into law you must, to that extent and to that extent, only, specifically make exceptions to the misuse doctrine, and that is the purpose of paragraph (d).” Hearings, supra, note 1, at 161-162.
“. . . The last paragraph of this section provides that one who merely does what he is aúthorized to do by statute-is not guilty of misuse of the patent.” H. R. Rep. No. 1923, supra, note 3, at 9.
The 1952 Act carried out this purpose by providing:
“(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.” 35 U. S. C. § 271 (a).
Although there was no statutory provision defining infringement prior to 1952, the definition adopted is consonant with the longstanding statutory prescription of the terms of the patent grant, which was contained in § 4884 of the Revised Statutes as follows:
“Every patent shall contain a short title or description of the invention or discovery, correctly indicating its nature and design, and a grant to the patentee ... of the exclusive right to make, me, and vend the invention or discovery throughout the United States . . . .” (Emphasis supplied.)
This provision is now contained without substantial change in 35 U. S. C. § 154.
This point was made by Mr. Rich in the course of the hearings as follows: “If the part he is supplying is in substance the very thing which was invented, it seems to me personally, that he is an infringer, and he should not be let off on some little technicality that there is something minor in the whole apparatus that he is not supplying.’’ Hearings, supra, note 1, at 153. (Emphasis supplied.) Of course, as Mr. Rich well knew, “the very thing” which is invented in any combination patent is the combination itself and not any unpatented component part of it — in this case, the whole convertible top, not the fabric used in it.
Cf. Mercoid Corp. v, Mid-Continent Inv. Co., 320 U. S. 661, 679-680 (Jackson, J., dissenting on other grounds).
See Transcript of Record, pp. 29-33, Cotton-Tie Co. v. Simmons, 106 U. S. 89.
I am perfectly content with the dissenting opinion’s denomination of our interpretation as a “reconstruction” of the cases in the sense in which I understand that term to be applied in these cases. I regret to say, however, that we cannot claim this interpretation to be a “discovery” in the patent sense. It was anticipated by Judge Learned Hand’s citation of Wilson v. Simpson for the proposition that “[t]he [patent] monopolist cannot prevent those to whom he sells from . . . reconditioning articles worn by use, unless they in fact make a new article.” United States v. Aluminum Co. of America, 148 F. 2d 416, 425.
If the wording of the statutory provision be thought to leave any doubt on this point, the intent of Congress is made manifest by the fact that the position of the word “knowing” (originally “knowingly”) in subsection (c) was changed after the hearings for this very purpose. See Hearings, supra, note 1, at 13, 142, 164, 176.
67 Stat. 232, as amended, 15 U. S. C. §§ 631-651.
Bloomer v. McQuewan, 14 How. 539, 549-550; see United States v. Univis Lens Co., 316 U. S. 241, 251, and cases collected in General Talking Pictures Corp. v. Western Electric Co., 305 U. S. 124, 128, note 1 (dissenting opinion). This century-long mandate cannot be set aside because of what my Brother Hablan now conceives to be a more enlightened policy in this field of law.
See Mercoid Corp. v. Mid-Continent Inv. Co., 320 U. S. 661, 667-668, and eases cited.
Id., at 679 (dissenting opinion).