COXE, District Judge.

These two suits are for infringement of patents relating to hat renovating devices, and have been tried together as one action. The first suit involves reissue patent 16,608, dated May 3, 1927, for an iron supporting plate for hot water boilers, and patent 1,-678.295, dated July 24, 1928, for a supporting iron plate for boilers. In the second suit, patent 1,646,098, dated October 18, 1927, covering an attachment for hot water boilers, is alleged to have been infringed. All three of the patents describe an improved construction for mounting a heating plate on top of a vertical steam boiler, and particularly the positioning of the plate in such a way as to minimize space and take advantage of whatever heat radiation there may be at the top of the boiler.

The boiler, heating plate, and burner are all stock parts, and invention is assei’ted for the manner in which they are attached and their relative positions in the organization. In each patent, the heating plate and burner are located directly over and concentrically with the head of the boiler; there is also an additional burner underneath the boiler for the purpose of generating steam.

In the reissue patent 16,608, dated May 3, 1927, the heating plate is attached to the boiler by a cylindrical clamping band. Patent 1,646,098, dated October 18, 1927, on the other hand, shows clamps for engaging the rivet heads of the boiler. Finally, in patent 1.678.295, dated July 24,1928, there is a central stud for securing the heating plate to the top of the boiler.

It is ádmitted that there is no proof of infringement with respect -to the second patent 1,646,098, and counsel concedes, therefore, that the bill of complaint in the second suit (Equity 51-184) may be dismissed.

The claims of the two remaining patents relied on are Nos. 1, 2, 4, and 5 of the reissue *479patent 16,608; and Nos. 1, 2, 8,10, and 11 of patent 1,678,295.

The defenses are invalidity due to lack of invention and aggregation, prior public use, and noninfringement.

Claim 4 of the reissue patent 16,608, and claim 1 of patent 1,678,295, are fairly typical of the other claims, and read, respectively, as follows:

Reissue patent 16,608: “(4) The combination with a boiler and an iron holder including depending legs seated upon the top of the boiler and spacing the top of the holder from the top of the boiler, of a clamping device carried by the depending Rgs for detachable gripping contact with the boiler for holding the holder against displacement.”

Patent 1,678,295: “(1) The combination with an iron holder and a boiler, of means for supporting the iron holder in superimposed relation to the boiler to conserve space- and utilize heat arising from the boiler including a stud mounted in the wall of the boiler and secured to the iron holder.”

The prior patented art is well illustrated by the Flicker patent, 1,156,356, dated October 12, 1915, and the Bqnenati patent, No. .1,386,230, dated August 2,1921. The Flicker patent is for hat blocking apparatus for straw hats, and shows a vertical steam boiler with an underneath burner and a heating plate mounted on supporting legs on top of the boiler. No second burner is shown under the heating plate, but the specification states that “the heat from the boiler will in turn heat” the plate, and “the h.eat will be communicated to the rim of the hat which is held in proper position by the weight of the actuate irons.” The Bene-nati patent shows a heating plate for sadirons mounted on top of the boiler, to one side of the center, and heated by an underneath gas burner. The specification states that “the fuel of the boiler may also be used to heat a pressing table.”

Coneededly, these two- patents are- the nearest patent references to Markatos, and it is urged by the defendants that, in view of the disclosures made by them, it was not invention merely to move the Bonenati heating plate to a position directly over the top of the boiler. It is also insisted that it was old with Flicker and Benenati to utilize the- boiler heat to warm the heating plate, and that in any event there was no invention in the idea.

In addition to the patent references urged by the defendants, there has been testimony regarding an alleged prior use by one Rosenbaum, which, it. is contended, is -a complete anticipation of the patents.

The proof with respect to this alleged pri- or use consists principally of (1) memory testimony by the two Rosenbaums and Podolsky, a former employee of theirs, to the effect that for a considerable period prior to 1920 the Rosenbaum Company manufactured •and sold in quantity hat renovating machines-like Exhibit F, and that the plaintiff himself purchased such machines from the Rosenbaums as long ago as 1908 and 1910, according to the elder Rosenbaum, or “about 19.19' or 1920, 1921,” according to the younger Rosenbaum; (2) a drawing and a tracing (Exhibits B and E), both undated and bearing the name “E. Garcia”, and showing a machine substantially like Exhibit F; (3) a drawing of a circular heating plate (Exhibit G), dated March 16,1919, and containing the names in print of the Rosenbaum Company and “Geo. Beck,” a draughtsman of the Ro'senbaums; (4) memory testimony of one Saullas that he purchased Exhibit F from the Rosenbaums in 1920, and used it in his hat cleaning establishment in the Bronx until the business was transferred to Picciolo in 1928; (5) memory testimony by one Cannelis that ho purchased a machine similar to Exhibit F from the Rosenbaums in 1918.

In this entire mass of proof, there is-practically nothing to meet the very severe test of this circuit, requiring that “memory testimony” of an alleged prior use bo supported by contemporaneous records, verbal or structural, in order to be effective. A. B. Dick Co. v. Simplicator (C. C. A.) 34 F.(2d) 935, 940; Block v. Nathan (C. C. A.) 9 F. (2d) 311, 313.

According to the elder Rosenbaum, thousands of machines similar to Exhibit F were-manufactured and sold by the Rosenbaum Company commencing in 1914; and yet therearo no catalogues, circulars, advertisements, or contemporaneous records to prove the fact — nothing of any substance except the Garcia drawings (Exhibits D and E) and the Beck drawing, dated March 16,1919 (Exhibit G).

It is stated in answer that the books of the Rosenbaum Company have been destroyed and are no longer available; that Garcia has disappeared; and that Beck is dead. All that is left is the testimony of the Rosenbaums and their former employee, Podolsky, that the Garcia and Beck drawings were made before 1920, and that they illustrate the type of machine manufactured and *480sold at that time. Obviously, this is not the kind of corroboration required in this circuit. Furthermore, the Beck drawing, which is the only one bearing a date, contains measurements different from the measurements of the heating plate on Exhibit F, indicating, that the drawing could not have been used in connection with the heating plate of that particular machine.

The testimony regarding the Saullas machine (Exhibit F) was not satisfactory. The older Rosenbaum stated that the machine was sold to Saullas “about 1918 or 1920,” but he was so vague and uncertain in his other testimony that little reliance can be placed on his recollection of dates. Saullas,«on the other hand, was more definite'with respect to’ dates, and testified that he purchased Exhibit F for $45 from the Rosenbaums in 1920 during the time he was in business at West Kingsbridge road in the Bronx. On cross-examination, however, he became confused, and, in attempting to account for his movements since his arrival in this country in 1907, placed himself at West Kingsbridge road at a date subsequent to 1920. There is an obvious mistake in the stenographic minutes with respect to the time Saullas spent in Boston; but figuring the period at ten years, as shown by my trial notes, instead of thirty years, as appearing in the minutes, Saullas could still not have been at West Kingsbridge road until 1922.

According to Saullas, Exhibit F was moved to a new location in the Bronx about 1925, and in 1928 it was sold to Piceiolo. Piceiolo in turn disposed of the machine to the defendant, Benanati, and it was identified at the trial by Saullas, Piceiolo, and Benenati. The machine now bears no distinguishing marks to show either its origin or the period of its manufacture. Nor are there any receipted bills, checks, memoranda, or book entries to corroborate the testimony regarding the various transfers.

The proof relating to the Cannelis ma'ehine was much weaker than that concerning the Saullas machine, particularly as the machine itself was not produced at the trial. There were also discrepancies in Cannelis’ testimony which seriously impair his general credibility.

With respect to the assertion of the Rosenbaums that the market was flooded with machines like Exhibit F prior to 1920, I find it difficult to reconcile such testimony with the fact that the defendant, Benenati, applied for patent No. 1,386,230 on February 3,1920, and Markatos applied for his first patent on June 14, 1923; and I cannot believe that these applications would have been made without reference to the Rosenbaum machine if it had been in existence and was as extensively in use as the Rosenbaums have testified.

I do not think, therefore, that the Rosenbaum use has been satisfactorily proved. The Barbed Wire Patent, 143 U. S. 275, 12 S. Ct. 443, 36 L. Ed. 154; A. B. Dick Company v. Simplicator, supra; Block v. Nathan, supra.

The ease then necessarily falls back on the prior art, as disclosed by the prior patent references. And on that basis I do not believe that wWt Markatos did constituted invention. In each of the patents the dominating idea is to economize space by placing the heating plate over, and concentrically with, the top of the boiler. Manifestly this required no exercise of the inventive faculty, inasmuch as any skilled mechanic could readily have adapted the Benenati machine to accomplish such a result. Furthermore, the idea of taking some of the heat from the boiler to warm the heating plate was not new with Markatos, but was disclosed in the Flicker and Benenati patents.

I cannot see, either, that the method of attaching the heating plate to the boiler by means of a cylindrical band, lugs, or a central stud, as shown by the three patents, involved invention. Obviously, these were mere mechanical details which were well known and understood prior to Markatos, and required only mechanical skill.

The Markatos patents are also to be condemned as mere aggregations of old elements without producing any result due to their joint and co-operating action. Reckendorfer v. Faber, 92 U. S. 347, 23 L. Ed. 719; Hendy v. Iron Works, 127 U. S. 370, 8 S. Ct. 1275, 32 L. Ed. 207; American Chocolate Machinery Co. v. Helmstetter (C. C. A.) 142 F. 978; Grinnell v. Johnson, 247 U. S. 426, 38 S. Ct. 547, 62 L. Ed. 1196; Powers-Kennedy v. Concrete, 282 U. S. 175, 51 S. Ct. 95, 75 L. Ed. 278, decided by Supreme Court, December 15,1930. The only suggestion of unitary action in the Markatos machines is that the heating plate in its superimposed position derives some heat from the boiler; but that contention is largely specious, and I am satisfied from the testimony that the boiler heat is practically negligible in its effect on the heating plate. Indeed, the mere presence of the burner under the heating plate bears testimony to the fact that the plate itself can*481not properly function by means of the heat derived from the boiler.

Finally, there is little in the commercial history of the Markatos machines to support the claim of invention. Kurtz v. Belle (C. C. A.) 280 F. 277. The largest number of such machines sold in any one year was 90 in 1925, and the average for the three years 1926-1928 was 36, thereby indicating only a small measure of commercial success for the machines.

I therefore find that Reissue patent 16,-608, dated May 3, 1927, and patent No. 1,-678,295, dated July 24, 1928, are invalid for lack of invention; and the bills of complaint in both actions are dismissed, with costs.