HAND, Circuit Judge (dissenting).

This is an appeal from a judgment, entered December 11, 1957, which held valid and infringed claims 3, 9 and 10 of a patent issued to Osius on March 1, 1938, and which directed the defendant to account. No injunction issued, or could issue, after March 1, 1955, the date when the patent expired. The invention was for an improvement "in the art of mixing fluent substances, and relates more specifically to * * * disintegrating, mixing and aerating materials to produce palatable and beneficial drinks or the like." The disclosure is of two elements. The first is a glass "receptacle" to receive and hold fluids and other food products, while they are being mixed. It is a tall upright glass jar, whose contour in section consists of connecting segments of four circular "lobes." At the bottom is a "rotor," consisting of a revolving metal member fastened to a shaft that passes through the bottom of the "receptacle." The "rotor" has four "blades," arranged like a steamer's propeller; two opposite "blades" that "incline upwardly and outwardly," and two other blades "horizontally arranged." The "receptacle" with the "rotor" and its shaft rests upon a "hollow base," containing a motor that actuates the shaft, the "receptacle" being held from rotating upon the "base" by "upstanding prongs." The motor drives the shaft at very great speeds and causes the contents of the "receptacle" to be thrown outward against the walls of the "receptacle," to be thence forced upwards in four "vortices," and finally to fall back to the bottom of the "receptacle." By repetition of this process "pulpy or fibrous" materials can be broken up and intimately mixed with fluids. In cross section the "lobes" of the "receptacle" are "smoothly arcuate and substantially semi-circular in form being defined in all horizontal planes by arcs of circles which are substantially tangential to the axis of the receptacle, as may be readily determined by an inspection of Figs. 4 and 5, preferably, four in number" (page 2, lines 67-73). Claim 3 described the "lobes" as "edge-connected," as Figures 4 and 5 showed them. Claims 9 and 10 substituted for "edge-connected" the phrase: "annular series of smoothly arcuate lobes, * * * each defined * * * by arcs of overlapping circles * * * each of which circles passes closely adjacent" to the "axis" of the "receptacle." This phrase necessarily results in "edge-connected" lobes, if the perimeter of the "receptacle" is to be "bounded by them": that is, if it is not made up in any part of straight lines.

On the day before Osius filed his application for the patent in suit — March 13, 1937, — he filed an application for a design patent which resulted in Design Patent 104,289, issued on April 27, 1937. This was for an "ornamental Design for a Drink Mixer" and consisted of two figures, of which one was described as the "plan view," and the other, as a "side elevation." (The numbers "1" and "2" were patently misplaced.) The "side elevation" disclosed a "receptacle" with a supporting base, the same in appearance as the outside of the "mixer," disclosed in Figure 1 of the patent in suit. On January 3, 1933, Osius had procured an earlier design patent (No. 88,888), which disclosed a "drink mixer" the perimeter of whose "receptacle" was substantially identical with that of the second design patent and of the Figures 4 and 5 of the patent in suit. The first design patent did not disclose any "rotor" in the base.

Judge HINCKS and Judge WATERMAN believe that the patent was valid and that it was infringed and they are giving their reasons in the accompanying opinion which is the opinion of the court. In what follows I have stated the reasons that lead me to believe that the patent was invalid and that the judgment should be reversed and the complaint dismissed. What I say is therefore only a dissenting opinion.

A design patent, issued before a mechanical patent, or a mechanical patent, issued before a design patent, is part of the prior art, and a valid anticipation. Lein v. Myers, 2 Cir., 105 F. 962; Bradley v. Eccles, 2 Cir., 126 F. 945, 949-50; Roberts v. Bennett, 2 Cir., 136 F. 193; Electro Mfg. Co. v. Yellin, 7 Cir., 132 F.2d 979, 981; Cary Mfg. Co. v. Neal, C.C.S.D.N.Y., 90 F. 725. The Court of Customs and Patent Appeals has several times held that a mechanical patent will anticipate a design patent, In re Staunton, 35 F.2d 63, 17 C.C.P.A. 579; In re Dalton, 37 F.2d 420, 17 C.C. P.A. 826; In re Rutledge, 47 F.2d 797, 18 C.C.P.A. 1081, and has also held that when the disclosures are the same or when it requires no invention to combine a mechanical disclosure and that of a co-pending design patent, the mechanical patent will be invalid for double patenting. In re Hargraves, 53 F.2d 900, 19 C.C.P.A. 784; Application of Phelan, Cust. & Pat.App. 205 F.2d 183. See also H. C. White Co. v. Morton E. Converse & Son Co., 2 Cir., 20 F.2d 311. "Drink mixers" with "receptacles" containing "rotors" at their base, either hand driven or motor driven, had been common in the art for many years before Osius filed his first design patent in September, 1932. (Wilton, No. 1,166,568; Stainbrook, No. 1,420,774; Poplawski, Nos. 1,480,912 and 1,927,184; Gilchrist, No. 1,606,992.) Even butter churns had had rotors. (Miller, No. 96,604; Crofford, No. 1,131,336.) It seems to me plain therefore that the disclosure of a "drink mixer" in Osius's first design patent must be read to presuppose a "rotor." It is true that the second design patent did, as I have said, disclose a "rotor"; but, strictly speaking, that was not part of its "design" for a design must be "ornamental" (H. C. White Co. v. Morton E. Converse & Son Co., supra), and the "rotor" is at most times concealed. In any event I cannot imagine what else than a "drink mixer" with adequate mixing mechanism the first design patent could be thought to have disclosed. Moreover, although claim 10 describes the "rotor" as "comprising a pair of horizontal blades * * * and a pair of upwardly and outwardly inclining blades both said pairs of blades having cutting edges," "blades" with "sharp edges" had been disclosed in Warrick, 1,803,446 (page 2, lines 11 and 12), and there was no new feature in claim 10 except that one pair of "blades" should extend "upwardly." This feature of that claim was also apparently disclosed in earlier "rotors" (Crofford, No. 1,131,336); but it hardly seems necessary to say that the patent cannot rest upon such a detail alone.

So I submit all that Osius did was to substitute a lobular "receptacle" for one whose inner surface was interrupted by baffle ridges or ribs; and, indeed, that had been disclosed over sixty years before in Miller's butter churn (No. 96,604). I need not say that it did not require invention to take Miller's butter churn, actuated by a hand driven "rotor," and apply it to a different art, in which the speed of the "rotor" and the purpose of the combination were so different. Arguendo, I will assume that that might have supported a patent; but after Osius had filed his first design patent for a "drink mixer" in an art that showed many "rotors" motor driven, I cannot agree that there was left any room for the combinations claimed in claims 3, 9 and 10. I do not find it necessary therefore to pass upon whether the plaintiff is estopped by the file wrapper to assert that the claims cover the defendant's "drink mixer." I would reverse the judgment and dismiss the complaint, because the claims in suit were invalid.

HINCKS, Circuit Judge.

We think that the Osius patent in suit was valid.

Although the prior art was replete with devices to mix liquids, Osius was the first to achieve his stated objective, viz., an improved apparatus not only to mix liquids but also effective to accomplish rapid and thorough disintegration, mixing and aerating of solids, including pulpy and fibrous materials, and fluent substances to form a uniform and creamy blend in which the solid material is thoroughly emulsified. His application correctly designated the invention as a "disintegrating mixer for producing fluent substances." That it achieved this objective was found by Judge Smith and there is abundant evidence to support that finding. That the invention had great utility not only for mixing foods and drinks but also for many other purposes never theretofore achieved by a drink mixer was also found below on an abundance of evidence. The Waring "Blendor" which embodied the Osius disclosure became a common item not only of household but also of laboratory equipment.

There was also ample evidence to support the crucial determination that the difference between the subject matter of the Osius disclosure and the prior art was not one which would have been obvious to a person having ordinary skill in the relevant art. Lyon v. Bausch & Lomb Optical Co., 2 Cir., 224 F.2d 530, certiorari denied 350 U.S. 911, 76 S. Ct. 193, 100 L. Ed. 799, rehearing denied 350 U.S. 955, 76 S. Ct. 341, 100 L. Ed. 831. His essential elements were (1) the container having lobed side walls, (2) containing in its bottom an impeller or rotor described in considerable detail and designed to provide both horizontal and vertical circulation, (3) the blades of which were sharpened and thus adapted to achieve the desired disintegration. The cooperation of these elements operated to raise cylindrical columns of liquid, one in each lobe, all of which coalesced to form a common and exaggerated vortex at the central axis of the receptacle, down which the liquids passed carrying any solid particles into disintegrating contact with the whirling, double-functioning blades of the rotor at the base and thence into an upwardly propelled column for continuing recirculation. Thus these elements, although each old in the art, produced results in the combination never before achieved.

The Wilton patent, No. 1,166,568 (1916), lacked both the lobular receptacle and the rotor blades of Osius' specification. So also did Stainbrook, 1,420,774 (1922); Poplawski, 1,480,914 (1922); and Gilchrest, 1,606,992 (1922). The old-fashioned butter churn of Miller, 96,604 (1869), though disclosing as its receptacle a fluted, or lobed, receptacle and two pairs of wings or paddles on the dasher rod at the top and bottom of the receptacle, was essentially a separating machine rather than a mixer or blender: its elements were neither designed nor effective to accomplish the continuing circulation of its contents which brought the solid particles thereof into repeated contact with the disintegrating blades. Warrick, 1,803,446 (1928), it is true, disclosed rotor blades with sharp edges but its combination did not purport to cover the receptacle and even its blades lacked the detail specified by the patent in suit. Crofford, 1,131,336 (1915), also a churn, specified a dasher shaft having wings "which may be of any improved construction," and its body or receptacle, although showing concave surfaces at its four corners was lacking in lobes which were either edge-connected or defined by overlapping circles. And Crofford, no more than Miller, gave any inkling of Osius' combination. And the Osius design patent Des. 88,888 (1933) went no further than to show the lobular receptacle of the patent in suit; wholly lacking was disclosure of the cooperation between the lobed receptacle and the rotor described in the patent in suit.

It is easy now, with the aid of hindsight to say that it was the simple thought of a skilled artisan to transform the churns and drink-mixers of the prior art into the combination disclosed in the patent in suit. But the fact that so much time intervened between the disclosures of the prior art and the date of Osius' application is persuasive demonstration that the Osius disclosure was not obvious even to those skilled in the art. And that conclusion, as Judge Smith properly found, is fortified by the substantial commercial success attained by Waring Blendors which embodied the invention of the patent in suit. We conclude, therefore, that the claims in suit1 were rightly held to be valid and pass on to a consideration of the finding of infringement made below.

As to Claims 9 and 10, it is clear that the plaintiff is estopped by the file wrapper to assert that the defendant's device infringes. These claims, by the patentee's amendment of May 29, 1937, had been added to the application as Claims 29 and 30. As was stated in the discussion included in that amendment, they were a rewriting of application claims 4 and 18, both of which covered combinations in which the lobes of the receptacle were broadly and similarly defined. Claim 4 had been rejected because "Crofford shows substantially applicant's form of lobes." And both Claims 4 and 18 had been cancelled by the amendment of May 29th. But both Claims 4 and 18 as rewritten in said amendment disclaimed the broad definition of the lobes originally used and substituted a more limited definition of the lobes, viz., that each was "defined in all horizontal planes by arcs of overlapping circles * * *." (Emphasis supplied.) Thus the Patent Office history makes it plain that this limitation, which did not appear in Claims 4 and 18 of the application, had been added to the rewritten claims to overcome the rejection of Claim 4. Concededly, the horizontal arcs of the lobes of defendant's device define circles of diameter so small that they do not overlap and in this respect these claims do not read on the defendant's device. Under the doctrine of file wrapper estoppel the plaintiff may not assert a broader field of monopoly than that restricted by its amendment made to escape rejection. Consequently, we now hold that these claims were not infringed.

However, careful analysis of the file wrapper fails to show any estoppel as to Claim 3. This claim had been added to the application as Claim 22 by the amendment of May 24th before Claims 3 and 4 of the application, with their broad definitions of the lobes, had been cancelled: it was not a rewritten substitute for Claims 3 and 4 of the application which in amended form ripened into Claims 8 and 9 of the patent as issued. To be sure, its definition of the lobes was narrower than that of Claims 3 and 4 in that it specified that the lobes were "edge-connected." But that limitation was no broader than Claim 5 of the original application which defined the lobes as "end-connected": it is clear that, in context, "end-connected" meant nothing different from "edge-connected." It is true that Claim 5 was rejected and never thereafter rewritten. But the basis of the rejection was not, as in the case of Claims 3 and 4, an overbroad definition of the lobes which left them, patentwise, indistinguishable from Crofford: it was rather because "to operate the stirrer of Crofford through the bottom of the receptacle would not constitute invention in view of Poplawski" who had earlier disclosed a rotor in the receptacle operated through its bottom. In other words, Claim 5 was rejected because it was thought that in the combination the rotor — not the lobes — was too broadly stated. Nothing in the file wrapper indicates that any claim defining the lobes as "end-connected" or "edge-connected" had been rejected because that definition of the lobes was overly broad; or that the "edge-connected" definition now contained in Claim 3 had been inserted to overcome a rejection of any claim on that ground. Thus by cancelling Claim 5, with its "end-connected" definition of the lobes in combination with an overbroadly defined rotor, and simultaneusly adding Claim 3 with its "edge-connected" definition of the lobes in combination with a more narrowly defined rotor, the patentee gave up nothing with respect to the lobes and as to that feature of the combination is not subject to a file wrapper estoppel. Cf. C. Howard Hunt Pen Co. v. Radiant Point Pen Corp., 2 Cir., 135 F.2d 870, 873; New York Scaffolding Co. v. Whitney, 8 Cir., 224 F. 452, 463, certiorari denied 239 U.S. 640, 36 S. Ct. 161, 60 L. Ed. 482.

The defendant contends that its product, the Universal Mixablend, does not infringe Claim 3 in suit because the Mixablend receptacle lacks "edge-connected" lobes. Instead, its lobes are separated by flat planar surfaces in the wall of the receptacle. However, as Judge Smith found, this difference of shape causes no difference in the functioning of the lobes in the combination or in the results achieved. In the operation of both devices each lobe shapes an upwardly-rising spiraling column of the material to be blended and the four columns coalesce to form a common exaggerated vortex down which the material whirls to contact with the blades of the rotor. In the absence of an applicable file wrapper estoppel, we think that the shape of the Mixablend receptacle was a mechanical equivalent of that literally disclosed in Claim 3 and that the claim, reasonably construed, covers the Mixablend device. Sanitary Refrigerator Co. v. Winters, 280 U.S. 30, 50 S. Ct. 9, 74 L. Ed. 147; Musher Foundation, Inc. v. Alba Trading Co., 2 Cir., 150 F.2d 885.

We hold, accordingly, that Claim 3 in suit was valid and infringed. And the decree below, modified to conform with this holding, is

Affirmed.

WATERMAN, Circuit Judge, concurs.