This is the usual infringement suit brought by the Victor Company against the Cheney Company, based upon claims 42 of patent No. 814,786, and 7 and 11 of patent No. 814,848, both issued March 3, 1906, to E. R. Johnson, and assigned to the Victor Company. The District Court held that claims 7 and 11 were not infringed, but that claim 42 was valid and infringed. Both parties appeal.
[1] Passing by other questions, we have thought proper to devote our attention chiefly to the issue of infringement of claim 42. That disposed of, the issues under claims 7 and 11 give less trouble. Claim 42 is of that type which seems to be simple and clear enough as applied to the particular structure described and shown in the patent, but which becomes thoroughly ambiguous when application is sought to the variant structure of a future defendant. It is also of that type where, without distortion of any word beyond the common meaning, the language may be read upon defendant’s structure, but where many things warn against the breadth of construction necessary to such application. Since the case presents an unusually complicated instance of the typical difficulties, and since our conclusion is superficially— *446though we think not substantially — not in accord with some results reached in other courts, it seems fitting to discuss the issue more in detail than we commonly do.
In 1903 there were two classes of sound recording and reproducing machines. One, which may be called the Edison type, used a record of cylindrical form, and the stylus followed a spiral path around the surface of the revolving cylinder by reason of a positive mechanical feed which caused relative motion longitudinally of the cylinder between it and the stylus-carrying parts. The other, which may be called the Berliner form, used a fiat disc, upon the upper surface of which the stylus traveled in a spiral path. In reproducing, the stylus point would tend to remain in the prepared groove, and thus to cause the stylus and its attached parts to travel from the outside of the disc towards the center. Each form was provided with a diaphragm operated by the stylus and communicating with an amplifying horn.
Johnson devised a sectional horn, the preferred and illustrated form of which was adapted particularly for use in the Berliner machines. He filed his application February 12, 1903, upon a talking machine. In February,. 1904, using identical drawings and generally the specification of the first ■ application, he filed a divisional application directed to the amplifying horn. Both patents issued on the same day, the one based upon the original application being No. 814,786, and the one based upon the divisional application being No. 814,848. The structure is shown in the following sketch, which is Fig. 1 of the drawings of each patent:
*447Claim 42 reads as follows:
“A talking machine, comprising a tapering sound-conveyor, moans for attaching sound-reproducing moans to the small end thereof, and horn-coupling and supporting means with which the other end of said conveyor is movably connected.”
The defendant manufactures a form fully enough shown by the following sketch:
As we approach the question whether claim 42 may, consistently with its validity, have a reading broad enough to cover defendant’s form, we do so in an atmosphere colored by two unusual things. The first is that plaintiff declined defendant’s offer to submit its machine to plaintiff soon after it carne oh the market, so as to be advised whether plaintiff would consider it an infringement of any patent, but later brought and prosecuted an infringement suit substantially the same as the present one, yet, when that suit was about to be heard, voluntarily discontinued it without prejudice to a new suit, and some three years later brought the present action. In this course of conduct we do not find the estoppel which defendant urges; but plaintiffs do not commonly take such action in clear cases, and its presence here strongly suggests that the right to recover in the first suit was doubted by the plaintiff.
The other colorful thing is that this patent application was prosecuted by skillful counsel for nearly three years, through repeated rejections and through the'presentation and urging of about 100 varying claims, resulting in a final sifting by which 40 claims were agreed upon *448between examiner and solicitor as covering the varying aspects of the invention, all before any claim occurred to the solicitor which would reach defendant’s structure. Just as the case was ready for issue claims 41 and 42 were added. This' suit is not planted on claim 41, although it is broader than 42. We do not suggest that the applicant may not, at any time before issue, broaden his claims in any way justified by his disclosure and by the state of the art; indeed, matters which develop during the period of prosecution often demonstrate or call attention to the fact that earlier claims are not as broad as they should be; but such a course of conduct as here occurred strongly supports an inference that the claim thus added was intended only to reach some anticipated, possible variations of the general conception already described and claimed, rather than a distinct and largely inconsistent conception which had never so far been suggested. Only in the latter view can the claim reach the defendant here.
Returning to the patented structure, we see that its primary elements are three: (1) The stylus with its diaphragm and diaphragm frame, which, in some form, is drawn down to a central opening opposite to the center of the diaphragm and constituting the beginning of a conduit for the sound waves which have been produced by the diaphragm vibrations. These parts, grouped in this way, seem to be what the pat-entee means when he speaks, in specification and claims, of a sound box. (2) The sound conveyor or tube which forms' a continuation of the conduit and carries the sound waves away from the production point in order to reach the amplifying hom. (3) The amplifying hom itself. Johnson makes his conduit (2) of expanding tapered form, and thus causes elements (2) and (3) to constitute together one continuous amplifying hom. In addition to these three primary, he has two secondary, elements. These are: (4) Connecting means between the sound box and the small end of the tapered tube giving relative movability, whereby the sound box can be raised or lowered for replacing a needle or starting or stopping and without moving the tapered tube. (5) Supporting and connecting means applied to the joint between the large end of the tapered tube and the small end of the hom proper, whereby either the tube or the horn may swing horizontally, and yet the weight of both is carried, and the two are coupled together into a unitary hom.
In order to reach the defendant’s form, elements 4 and 5 must be considered to cover all known means of operative connection between (1) and (2) and between (2) and (3). Defendant attaches its sound box to the small end of its sound tube by a bayonet joint. There is detachability but no adjustability of any kind. No method of attachment has been suggested which would escape the claim, if this one does not. At the other end of the sound tube defendant, who uses the now familiar cabinet style, supports the horn by permanently and rigidly fastening to the cabinet top, depending therefrom, another sound tube which at its other and lower end rigidly carries the hom proper. The member which serves for coupling the two parts of the horn (if there is any such coupling at the movable joint), does not support the horn. Seemingly, any form of supporting the horn and the tube so that they *449effectively communicate, but with relative motion, would respond to the claim, if this one does. In substantial effect, plaintiff says that claim 42 is for “a talking machine comprising a tapering sound conveyor, carrying at the small end sound-reproducing means, and at the large end communicating with a suitably supported horn and having a jointed connection therewith.” With this — necessary for this suit- — construction, the claim reads absolutely upon Baynes and Jenson of the prior art, save that their sound tubes were cylindrical, and not tapered. We therefore meet the questions whether there was any invention in this mere change from straight tube to tapered tube, and whether claim 42 should be given that breadth of construction which can rest only on the proposition that there was invention in this mere change.
Upon these questions we have no precedent in any earlier decision upon this patent. The opinions of Justice Warrington, in the Chancery Division, and of the judges in the Court of Appeal (Graphophone Co. v. Ruhl), indicate that no great breadth was accorded to the English patent which has the same drawings as both the patents here in suit; but the question of broad invention, as we have stated it, was not discussed; indeed, the English patent contained no claim of such scope; its broadest claim was like 7 of 814,848. In the Lindstrom Case, 279 Eed. 570, Judge Learned Hand states the question broadly enough and concludes that there was invention, but though claim 42 was sued upon and infringement thereof was found, yet defendant’s machine there responded to several other claims and would have infringed claim 42, even though construed narrowly enough not to reach the defendant here. If is fairly consistent with what Judge Hand says to conclude that he had in mind, not the mere change from straight tube to tapered, but that change associated with Johnson’s chief declared object — a continuously ampli fying horn from sound box to mouth. In the Wan-namaker Case, 275 Eed. 448, claim 42 was also sued upon and was found valid by Judge Augustus Hand; but here, again, several other claims were infringed, and the validity of the claim to the tapered tone arm, in combination with improvements at both ends which Johnson devised and which that defendant used, was the real question involved.
For the purposes of this opinion at least, we will assume that there was invention broadly in this mere change, and that Johnson would have been entitled to a claim like the one we have supposed. It does not follow that claim 42, as issued, was intended to have, or can receive, this construction. Here, again, we have no precedent in the previous litigation. The claim has received no special attention and has not been applied, except in cases where there was no question of infringement, beyond that involved in defining “tapering sound conveyor.”
In determining the scope, intended or appropriate, we cannot overlook Cannavel.1 Our foregoing assumption of validity implies also *450that Cannavel is not a complete anticipation; but he has a bearing on the scope. He used an Edison, rather than a Berliner, machine; but this cannot be controlling, since the Johnson specification does not suggest that his invention fails to reach both classes, and many of his claims, including 42, are as appropriate to one class as to the other. Cannavel showed the.complete sound box of Johnson, consisting of a stylus, a diaphragm, and a diaphragm frame drawn in back of the diaphragm so as to leave a small central opening opposite the diaphragm center. He then conducted the sound away from this central opening through an expanding taper tube toward the horn. This tube turns and extends parallel to the diaphragm a substantial distance beyond the diaphragm edge, but it is relatively short, it is made integral with the primary sound box and as a development thereof, and Cannavel calls it a diaphragm box. This first tube (<? in the German, cl in the French) is then attached by slip connection to a second tube (g in the German, e in the French) which continues the progressive taper expansion. At the other and larger end this second tube enters the basé member of the horn proper (Tin the German, f in the French), where it is pivoted, and through which the progressive expansion of the sound waves continues. Cannavel distinctly discloses, by his specification and drawings, the same meritorious thought which is at the base of the Johnson invention, as it is now claimed to be formulated in claim 42, viz. that the expansion in the sound tube should continue in unbroken progression from the immediate vicinity of the diaphragm on through into the main horn, and that there should be a jointed connection between the sections of this expanding horn which would permit the sound box to have the necessary play while the horn itself was otherwise supported.2 Cannavel’s c constitutes Johnson’s sound box and sound tube combined, save that the tube is so short that it may be thought dominantly a sound box only. Cannavel’s g is Johnson’s tapering sound tube movably connected with the horn at the large end and carrying sound-reproducing means at the small end, save that it is so short that its coupling function may be thought to dominate its function as a tube (Cannavel calls it “a short tube which constitutes a ball joint”); but it was tapered, it was a sound conveyor, and it was as long as necessary to reach from the sound box to the horn. However, we pass Cannavel by with the conclusions that, conceding invention in lengthening his intermediate tube coupling member, the field is narrow, and that, where we find this tube claimed in combination with other novel elements which Johnson had devised, the presence in the claim of the latter creates limitations which cannot be minimized by the thought that the tapered tube was a revolutionary invention.
Referring to the small end of the tapered tube, the claim calls for sound-producing means and the means for attaching the latter to the *451tube. If in plaintiff’s machine we substitute defendant’s means for attaching these two parts, the machine becomes inoperative; and this is sometimes taken as the test of equivalency. It is not a true test, to disprove equivalency, because the iuoperativcness may be overcome by compensatory changes at another place which may be within the skill of the ordinary mechanic, and we think that would he true here; hence infringement is not thus escaped by the difference at this point, and we see no reason for limiting “means for attaching” so as not to include defendant’s bayonet joint.
Coming to the large end of the tapered tube, we find that the structure of the patent provides a curving arm extending out and up from the main frame. This arm carries, rigidly attached and extending therefrom, a horizontal bracket, J, in the form of a flat-topped ring, 9 with an annular flange rising and a sleeve, 6, depending therefrom. The ring also carries a transverse central bar. The upper surface of the ring supports and carries the main horn positioned by the flange. The large end of the tapered tube enters this depending sleeve, which thus serves as a coupling, and is supported there by a pivot post which, in turn, is supported by the frame arm, but the tapered tube is not supported by the coupling. In many places the horn is considered as a complete unit, with two sections, but in this claim Johnson clearly differentiates between the tube and the horn, and when he says “horn” he means what he sometimes calls the “horn proper,” of main horn. We thus find a group of means (arm and ring, with flange and sleeve) specially devised by Johnson, which constitute a combined coupling between the horn sections and support for the large one, and which, when united in composite form, constitutes one means for both functions. We think the fair interpretation of claim 42 calls for such composite unit, though its form might be much, varied. Six prior claims had specified means for the coupling and means for the supporting functions; some of them very specifically and some of them broadly. When Johnson wanted to call for all means which would couple or all means which would support, he knew how to do so. In claim 6 he said, “said horn and tube being independently supported.” In claim 10 he said, “said horn and tube being supported to move.” The language of claim 11 aptly describes the coupling and supporting functions with the scope which plaintiff now seeks to give to claim 42. Claim 41 calls, by Implication, for the supporting function in the broadest way. The language of claim 1, omitting the sound box connection limitation, was admirably suited for the construction now claimed for the very different language of 42, which specifies “horn-coupling and supporting means.” 3 This seems to 11s, as we have said, to imply the conception of a means, beyond the mere frame of the machine, which, as a composite element or as a group of elements, should both support the horn and couple it and the tube. Defendant does not have any such *452element, unitary or compound. Its bom, if the hom extends back to this point at all, is of wood and supported by the wooden cabinet top or frame from which it depends, and is held there by an ordinary cabinetmaker’s glue joint. The large end of the taper tube rests indirectly upon, and is supported by, the same top or frame. The sleeve or coupling member (if coupling there is in the patent serise) also rests upon the same frame member. It is not supported by, nor does it in any degree, directly or indirectly, support, the main horn, though it does immediately support the tapered tube. We cannot find this “horn-coupling and supporting means” in defendant’s structure. The same result will follow if the call of the claim is thought to be for means for supporting and coupling both tube and hom.
There is another difference which is not clear as a matter of words, but is substantial and vital as a matter of substance. The claim calls for a “coupling” between the two parts of the hom. This requires that the two should come together so that they can be coupled. Johnson intended that the two. parts of this horn, coupled together, should constitute one amplifying horn, without substantial lack of continuity in the amplification. This will be further pointed out. In defendant’s sound tube we take the step by step enlargement (58-38) to be the equivalent (for the purpose of claim 42) of Johnson’s unbroken taper, and this brings substantially progressive enlargement until the passage has curved downward and has come to tube 73. Here there is a reduction in the cross-section area of nearly 40 per cent. At the bottom of 73 (39) there is a change from round to square form and consequent enlargement which approximately compensates for the 40 per cent, constriction above; then the passageway makes a square turn through a cubical chamber with first an' enlargement and then a further constriction of about 15 per cent, in passing through what Cheney calls his mechanical throat. Then, and then only, comes the other and larger section of the amplifying horn. The net result is that from the reproducing means the passage is continuously amplifying for a certain distance, then it is very substantially constricted and turns a square corner, all for a distance substantially the same as the length of the first tapered tube, and then only is permitted to expand more freely. This treatment is in the teeth of the teachings of the patent, and upon theories antagonistic to anything which can happen in the structure shown by the drawing.
The specification continually points out the advantages of the invention upon which a monopoly is sought. Collating these statements and omitting those which refer to subordinate features not involved in claim 42, we find:
“By locating the small end of the horn in this manner so that the sound-conducting tube or horn flares outwardly practically from the sound box, I have found that it allows the sound waves to advance with a regular, steady, and natural increase in their wave fronts, in a manner somewhat similar to that of the ordinary musical instruments, thus obviating the well-known disadvantages due to long passages of small and practically constant diameter. * * * It is also desirable to avoid abrupt turns in the sound-conducting tube or passage. * * * It is therefore the object of my invention to provide a talking machine with an amplifying hom meeting these requirements. * * * I provide, in effect, an amplifying horn that extends, *453practically, from the sound box. * ® * It consists of two sections, one of which is the tapering, hollow sound-conducting horn * * * mounted upon the machine, while the other section is the * * * horn proper. •5 * =■* The advantage of this is that I secure the requisite length of a constantly flaring or tapering horn which gives the desired result in quality and volume of reproduction. * * * The horn proper forms only a portion of the sound-conducting tube. * !l‘ * I have avoided to the greatest degree any abrupt turns. * * * 1 have produced, in effect, a sectional horn, tapering from end to end.”
In the progress of the application through the Patent Office the applicant made repeated arguments and discussions pointing out the advantages of his invention and the distinctions over the art cited. Every instance of these claims for merit or for invention, whether staled in the specification or in the arguments, is in such, terms as to exclude defendant’s construction. With this history, the alleged broad language of the claim should be very clear to justify finding infringement.
We have little hesitation in saying that defendant’s hom proper, as that part is intended by the specification, does not extend up through the tube 73 to the cabinet top to be there coupled to the tapered tube, nor in also concluding that the tube 7,f-does not itself constitute the coupling member which unites the two sections of the horn. The 10-foot section of an ordinary gas pipe which is interposed between two other similar sections and fastened to both truly enough couples them together, but it is not commonly spoken of as a coupling; on the contrary, it is a spacer which holds them apart; and an interposed member or element which destroys the theory of operation and of advantage claimed for the invention cannot be that coupling which the patent calls for in order to carry out the invention.
Previous decisions do not throw much light on the question of infringement. The breaks in progressive amplification have been, or have been said to be, unsubstantial, and the limitation to “horn-coupliñg and supporting means” has never been interpreted. Infringement of claim 42 has not been essential to justify any injunction that has been granted; it has never been worth while to determine its scope carefully.
These considerations require a reversal of the decree and a dismissal of the bill as to patent No. 814,786.
[2] As to the other patent, No. 814,848, where the court below held there was no infringement, the views already stated require an affirmance. Claim 7 calls for an amplifying horn which is — “comprising” — a continuously tapering tube with a joint between the two parts thereof. Claim 11 does not, in set words, require that the horn shall be a continuously tapering tube, but it describes the horn as “a tapering curved tube,” and this reference and description are to the tube as a whole, and not to any part. Such description does not aptly apply to a tube the central one-third of which is not tapered, but is parallel-sided and is very substantially constricted. As to these two claims, the decree is affirmed.
French, of May 4, 1901, German, oí 1901, filed February 28. Cannavel has not been mentioned in .previous decisions, save in the Ruhl case. The French drawing seems to indicate horizontal as well as vertical motion at the tube-horn joint, but it is not described. Any effort to carry the Johnson invention bads of Cannavel is not substantial.
He says (French patent): “The characteristic and essential point of my invention is that the successive channels <?i, e, and f widen after leaving the [sound bos] orifice, <?2. * * * In no case does there exist in one point of the channel a part not widening. * * * This point is very important, and this employment of a channel which widens in a continuous manner from the orifice,” etc.
Plaintiff’s witness Hunter describes the difference between claims 41 and 42 by saying of 42 that "it is specifically stated that the horn coupling also aets as a supporting means with which the other and larger end of the tapered sound conveyer is connected, whereas in claim 41 the coupling is not slated to act also as the ‘’support.”