DAVIS, Circuit Judge.

This is an appeal from a decree of the District Court in which it was determined that the Johnston "method" patent No. 1,388,434, the Johnston "product" patent, No. 1,435,306, and the Belmont patent, No. 1,611,875, were valid; that the defendant-appellant had infringed claims 1, 2, 4, 5, 10, 14, 17, and 19 of the Johnston method patent, claims 3, 5, 6, and 9 of the Johnston product patent, and claims 1, 2, 3, 4, and 6 of the Belmont patent; and that the defendant was liable to the plaintiff for damages.

The two Johnston patents were held to be valid by this court in the case of Steel Tubes, Inc., v. General Tube Co., 61 F.2d 475, 476, and by other federal courts in the cases of Elyria Iron Steel Co. v. Mohegan Tube Co., Inc., et al., 7 F.2d 827, 830 (C.C.A.2); Steel and Tubes, Inc., v. Greenpoint Metallic Bed Co., Inc., 37 F.2d 172 (D.C.E.D.N.Y.); and Steel Tubes v. Jackson Tube Co., Inc., 42 F.2d 760 (D.C.E.D.N.Y.)

In view of our opinion in the General Tube Case, supra, it is sufficient to say here that the Johnston patents involve the manufacture of butt-welded, thin gauge tubing. There is no need to repeat here what we said there with regard to the prior art or the contribution to the art by Johnston.

The defendant does not contest the validity of the Johnston method patent and there has been no evidence that persuades us to change our opinion from that expressed in the General Tube Case. The defendant does, however, present several arguments to the effect that the Johnston product patent was invalid. Since these points were not discussed in the General Tube Case, we will briefly consider them here.

The defendant contends that the Johnston product patent is invalid for lack of utility. This argument is untenable in view of the improvements in strength and physical structure of the weld, and the superiority of the Johnston product over that produced by the prior art.

The defendant further contends that the product patent is void for double patenting and that it was merely an attempt by the plaintiff to prolong the monopoly, but these contentions are not in accord with the facts. Both product and method claims were incorporated in the original application for a patent filed by Johnston on June 9, 1919. The Patent Office required that these claims be segregated and ruled that the product claims should be considered "in another division." The patent attorneys complied with this requirement and before the method patent had been issued filed application for the product patent in the other division. The method patent was granted on August 23, 1921, and the product patent on November 14, 1922. The defendant says that the patentee should have resisted the requirement of the Patent Office as invalid on the ground that the method and the product involved but a single invention. However, and even though the applicant may appeal from a ruling of the Patent Office, he cannot "justly be blamed for acquiescing in a command by lawful authority, much less can he properly be made to suffer loss by obedience." American Laundry Machinery Co. v. Prosperity Co. (C.C.A.) 295 F. 819, 821. Furthermore, as was said in that case: "It being `difficult, perhaps impossible,' to lay down general rules determining when improvements should be embraced in `one, two, or more' patents, discretion must be left to the Patent Office on this `nice and perplexing question.'" The case of Wirebounds Patent Co. v. Saranac Automatic Machine Corporation (C.C.A.) 65 F.2d 904, relied upon by the defendant, is not controlling in this case. In that case the patentee waited seven years after the original patent had been issued before filing certain method claims. These claims were held invalid on the ground that the applicant was guilty of laches in waiting so long.

The final contention of the defendant on this point is that the product patent is invalid because of public use for more than two years prior to the filing of the original application. To establish this fact the appellant relies upon the testimony of Mr. Zalmon G. Simmons, president of the Simmons Company, which was given in the Mohegan Case, supra, and incorporated in the present case by stipulation. This testimony reveals that Mr. Smith, a representative of Johnston, demonstrated to him several small pieces of tubing manufactured by the Johnston process without, however, disclosing the process. These demonstrations were made preliminary to the execution of a contract with Mr. Simmons in which he was given exclusive right to use these inventions in the manufacture of bedsteads and allied articles. The process was disclosed to Mr. Simmons in the contract which bound him to secrecy. The showing of these pieces of tube to Mr. Simmons and the demonstration of their fitness for his use without disclosing the process did not amount to public use. The defendant did not bear the burden required of him in establishing prior public use for more than two years prior to the application for the patent. The Barbed Wire Patent, 143 U.S. 275, 285, 12 S.Ct. 443, 450, 36 L.Ed. 154.

The defendant has been manufacturing welded tubing with direct current in some instances and with alternating current in others. The plaintiff alleges that claims 1 and 2 of the Johnston method patent were infringed by the defendant's direct current operations and that claims 4, 5, 10, 14, 17, and 19 of the same patent were infringed by the defendant's alternating current operations. We shall consider the alleged infringement by the alternating current operations first.

As we said in the General Tube Case, supra, the Johnston invention involved "a new correlation of the variable elements of speed, current, and pressure," or as Judge Hough of the Second Circuit Court of Appeals in the Mohegan Case, supra, said, the correlation of "time or speed of movement, heat both as to kind to be applied and area to be heated, contact of butts as to accuracy and pressure as to amount." The court in the latter case defined the correlation of these elements as follows: "To move the material in very accurate register at the seam cleft, as fast as the material could be heated to the welding point just at the cleft, in time to be subjected to a pressure that would cause practically no extrusion of metal in burr."

The defendant insists that it does not use the Johnston method, but rather uses the Parpart method and a Parpart machine which was the established practice prior to the improvements introduced by Johnston. The fact that the defendant might be using a Parpart machine is not conclusive on this point. Both Judge Hough and we said that the Johnston method could be practiced on a Parpart machine "after a reorganization not mechanically vital." The defendant does, however, point to a variation in the method it employs from the method generally practiced under the Johnston patents. It alleges that the axis of the pressure rolls on its machine is from 1/16" to ¾" (usually 3/8") beyond (that is, in the direction in which the tube moves) the axis of the electrodes.

The defendant alleges that this placement results in a weld accomplished by an accumulation of electrical impulses rather than by a "single shot" weld of the Johnston method. In support of this contention it points out that when the tube is traveling at a speed of 30 feet per minute and a 60 cycle alternating current is used that 1.06 impulses cross the seam cleft as the tube travels 1/16"; 7½ impulses cross as it travels ¾" and 15 impulses as it travels ¾". Thus as the offset between the pressure rolls and the electrodes increases from 1/16" to ¾" the electrical impulses that cross the seam cleft increase from 1.06 to 15.

The question, therefore, is whether or not the defendant escapes infringement of the Johnston method patent by the above practice. Does the above placement result in a single shot weld or does it result in a weld by the accumulation of electrical impulses?

If the offset above mentioned is 3/8" then it takes 1/16 of a second for the center line of the portion heated by each electrical impulse to pass from the axis of the electrodes to the axis of the pressure rolls. Does the fact that 6½ other electrical impulses cross the seam cleft during that 1/16 of a second affect the tubing in such a way that by the time the portion of the tubing heated by the first impulse reaches the pressure rolls, the effect of that first impulse cannot be said to be the prime cause of the weld at that point? The fact that most of the defendant's product exhibits the stitch-like effect is strong evidence that the weld produced by the defendant's process is a single shot weld.

The evidence further indicates that the defendant's process uses the correlation of speed current and pressure peculiar to the Johnston method, and there is nothing in the Johnston method patent that limits the method therein described to a process in which the electrodes and the pressure rolls were in exact vertical alignment. On these facts the District Court was correct in finding that the defendant infringed the claims of the Johnston method patent.

The second question is whether or not the defendant's direct current operations infringe claims 1 and 2 of the Johnston method patent as the plaintiff alleges.

Claim 1 which may be regarded as typical of both reads as follows: "1. An improved process of butt-welding thin gage tubing, which consists in passing a sheet metal blank of excess diameter through a throat formed by a plurality of rollers grooved to conform in size and shape to the exterior contour of the welded tube and held rigidly against yielding, so as to effect an upsetting of the blank edges and a reduction of the diameter of the blank as it passes through said throat, said rollers collectively embracing and supporting the blank throughout substantially its complete circumference and being arranged so that their line of contact lies in a single plane, causing said blank to progress continuously and steadily through said roller throat and simultaneously passing a current of relatively large amperage across the joint between the edges of the blank being butt welded, whereby concurrently with the heating and softening of the edges of the blank the latter is swaged and compressed to the size of the roller-pass, and welding thereby effected."

The defendant contends that the Johnston method is limited to the "single shot" weld, and since there are no "shots" in direct current, that these operations could not possibly infringe any of the Johnston claims.

The patent, however, does not limit itself to alternating current, but merely indicates a preference for it. On page 3 of the specification it is stated that "the electric current supply be approximately uniform; so far as I am at present advised it should best be alternating current. * * *" Neither claim 1 or 2 specified that either direct or alternating current should be used. Claim 1 says: "simultaneously passing a current of relatively large amperage across the joint," and claim 2 says: "passing a heavy electric current through and across the seam cleft."

Thus, though the "stitch weld" and "single shot" weld have been emphasized in the various court opinions dealing with this subject, the Johnston method is not expressly limited to alternating current, or to a method which would produce a product having the recurrent "stitched" appearance. As was said in the Mohegan Case, the General Tube Case and above, in this opinion, the Johnston invention involved a new correlation of the several "elements of speed, current and pressure" and if the Johnston correlation is used the patent is infringed. Since the defendant used the Johnston correlation of speed, current, and pressure in the direct current operations, and since the Johnston method patent does not limit its application to the use of alternating current, the District Court was correct in finding that claims 1 and 2 of the Johnston method patent were infringed.

The plaintiff does not contend that the product of the defendant's direct current operations infringe the claims of the product patent, but it does contend that the product of the defendant's alternating current operations infringes claims 3, 5, 6, and 9 of the Johnston product patent.

Claim 3 is typical of these claims and reads as follows: "As a new article of manufacture thin-walled electrically buttwelded steel tubing, wherein the abutting walls are united by a substantially continuous weld presenting recurrent variations in metal texture in closely-adjacent narrow zones synchronizing with varying applications of welding current."

The defendant admits that in so far as its product exhibits the "recurrent variations" it does infringe the above claims, but alleges that much of its product does not exhibit these variations, and therefore does not infringe.

Though the method patent is not expressly limited to a process which would produce tubing having recurrent variations in the weld, all of the claims of the product patent describe the product as having these variations. Therefore, it is evident that the claims of the product patent were infringed only to the extent that the product of the defendant exhibited these recurrent variations.

The last patent involved in this case is the Belmont patent, No. 1,611,875. This patent "has to do primarily with the smoothing out of the weld so as to obliterate any burrs that may have resulted from the welding operation." The aim of the patent is achieved by passing the newly welded seam, while still hot, between "a pair of rolling elements" having "relatively fixed axes." These rolling elements, one "being inside the tube * * * and the other outside," co-operate "to squeeze the welded seam between them * * * and thereby smooth out any internal or external burrs on the tube." The device is "adapted for use with the so-called continuous welding machines." The roller which is inside the tube is mounted on a mandrel having one or more supporting rollers. The mandrel is held in position by a rod extending back and beyond the welding electrodes and is fastened by a suitable bracket through the open seam cleft. The roller which smooths down the exterior burr "may be secured to the welding apparatus or any suitable support."

This patent was held to be valid in the case of Steel Tubes, Inc., v. Jackson Tube Co., supra, and by the District Court in the case at bar. The defendant, however, argues that this patent is invalid for the reason that the device described in the patent was not only well known to the prior art, but was also anticipated by several prior patents.

Both the Kellogg patent, No. 278,340, and the Lloyd patent, No. 768,534, describe mandrels with rollers to be used in conjunction with external rollers in the formation of metal tubing, but these patents did not suggest that the machines described therein could be used as a roller "for operating on the weld to smooth down the interior burr, a roll on the exterior of the tube opposed to said roller and operating on the weld to smooth down the exterior burr." Neither did the Sokol patent, No. 249,080, which in mechanical equipment is perhaps nearest to Belmont. He did say in describing a tube welding machine for making longitudinal seams in large diameter tubes that the rollers behind the hammer as in figure 6 are intended to shape the overlapped seam welded by the hammer "smoothly and according to profile," but Sokol did not have the conception of Belmont and in his claims he does not mention the disclosure which is claimed by Belmont as the very essence of his invention. We agree with Judge Schoonmaker, who said of the Sokol patent: "We find no disclosure which structurally or in function corresponds to the Belmont apparatus." It may be true that an expert, after Belmont had made his contribution to the art, might go to the art and select one thing from one patent and another thing from another patent and from many bring together a combination which would produce the result achieved by Belmont. So could it be done in almost every other patent except purely pioneer patents, but the Belmont patent described a new combination of old elements which produced a new result and is valid. Otis Elevator Co. v. Interborough Rapid Transit Co., 222 F. 501 (C.C.A.2); Barber v. Otis Motor Sales Company, 240 F. 723 (C.C.A.2).

We agree with the learned trial judge again when he said: "There seems to be no question but that the Belmont patent is infringed by defendant's structures. It is evident from the drawings of defendant's machine, Exhibits 30a, b, c, d, that the defendant's burr rolling device is constructed substantially in the same way and operates to produce the same results as Belmont."

The decree of the District Court is affirmed.