The sixth and last reference relied on by the Board is the Hoffmann patent. This patent discloses a method for the purification of tetrachlorophthalic acid and its anhydride. No mention is made of TBPAA. The ferric iron mentioned is obviously present because of the use of ferric chloride as a chlorination catalyst in the preparation of tetrachlorophthalic acid. TBPAA is generally not prepared in the same manner * * *. However, Hoffmann does not state that ferric iron is responsible for the difficulty in the formation of polyester, and, even if the suggestion is assumed, there is no teaching that a limited amount of ferric iron can be present. Furthermore, Hoffmann does not, in any way, suggest the special significance of TBPAA.

It is therefore apparent that none of the references, alone or in combination, suggest that polyester resins of remarkable flame retardancy can be prepared by utilizing TBPAA that has not above 5 parts per million of ferric iron associated therewith. The references relied on are not concerned with the problem of esterifying TBPAA, do not offer a solution, and are not aware of the special significance of TBPAA in the fire retardant field.

Claims 6, 7, 8 and 12 while generally in the form of "product" claims contain references to the "reaction product" (cl. 6-8) of "glycol and a component of the group consisting of saturated dicarboxylic acids and anhydrides" or to "a mixture of cross linking agents *** resulting from condensation of about one mol of glycol with about one mol of dicarboxylic acid ***." All of these claims contain the essential and critical limitation as to the ferric iron impurity in TBPAA not being above about 5 ppm. They thus embrace the essential elements of appellants' invention and appear to be little more than a permissible alternative way to claim their invention which resulted from a "surprising discovery" of the effect of the ferric iron impurity in the reaction for producing the desired fire-retardant resin. I therefore find them to be properly allowable claims.

It but further begs the question to find TBPAA as a known material in the prior art. It is the control of the ferric iron impurity therein which appellants related to the problem of utilizing TBPAA in polymerization reactions to produce fire-tardant resins. There is no question but that the art totally fails to teach this concept. The process and the product produced are clearly novel in this art. Thus, the only issue is whether under the tests of section 103 they are obvious.

In referring to the requirements of section 103, the Supreme Court in Graham v. John Deere, 383 U.S. 1, 148 USPQ 459 (1966) stated:

The emphasis on nonobviousness is one of inquiry, not quality and, as such, comports with the constitutional strictures.

While the ultimate question of patent validity is one of law, A. & P. Tea Co. v. Supermarket Corp., supra, at 155, 87 USPQ at 307, the § 103 condition, which is but one of three conditions, each of which must be satisfied, lends itself to several basic factual inquiries. Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved.

Against this background, the obviousness or nonobviousness of the subject matter is determined. * *

The present ex parte record is not as fully developed here as it should be to permit a more accurate determination of the present issue. The only evidence directly bearing on the existence of the problem and its solution by appellants is found in their specification. The examiner has cited no art bearing directly on the issue, except as we ignore the "invention as a whole" requirement of section 103 and reconstruct the invention on a piecemeal hindsight basis. Even when this is done, the art totally fails to make obvious the critical limits of the ferric iron impurity claimed by appellants. To bridge this gap, the examiner "presumes" that one of ordinary skill in the art of manufacturing fireretardant resins by the use of TBPAA would use chemically pure rather than commercially available TBPAA. Even here, he furnishes no "evidence" as to what would be obvious to one of ordinary skill in this art.

It seems clear to me that the intent of Congress expressed in the Patent Act of 1952 was to grant patents on an invention unless one of the statutory grounds for refusal was established. I do not consider the unsupported "presumption" of an examiner sufficient to establish obviousness under section 103 as interpreted by the Supreme Court in the Graham and Adams cases.

Judge Kirkpatrick's concurring opinion states:

The difficulty with the appellants' case now before us is the lack of adequate evidence to bring his application within the rule. There is nothing to show that the "problem" which the appellants claim to have solved is anything other than the perennial problem present in every industry, namely, how to improve the quality of the product or to make the process of manufacture more efficient. Certainly the mere fact that an applicant for a patent has succeeded in making an improvement in product or process is never of itself evidence of nonobviousness. It must appear that what he has done is not within the skill of the ordinary practitioner of the art. Here nothing has been offered to show that any problem other than the problem of how to improve the product and process existed.

As above pointed out, by disagreement is that I find in the specification "evidence," which in my opinion is far more persuasive than the "presumption" on which the examiner based the rejection. Briefly, it seems to me that an applicant should not be required to produce more "evidence" than was here presented until the examiner has overcome the prima facie case of patentability which I find in the present specification. Only after a properly supported rejection has been made is the burden of further proof properly placed on an applicant. It is a fact of which judicial notice is properly taken that this court, as well as other courts, as in Adams and in Eibel, have found patents valid when directed to nothing more than improving the quality of a

product or making the process of manufacture more efficient. The fallacy is that how one improves quality of a product or efficiency of a process frequently involves unobvious patentable improvements, and this is as it should be within the terms of 35 USC 101 which specifically includes "any new and useful improvement" on processes, machines, manufactures or composition of matter.

I would, therefore, reverse the decision of the board.

386 F. 2d 466; 156 USPQ 85

RICHARD F. DYER v. FREDERICK C. FIELD, JR. (No. 7815)

PATENTS

1. INTERFERENCE-INTERFERENCE IN FACT

Count limitations need not be set forth in haec verba in disclosure relied upon, it being sufficient if specification, considered in its entirety, is so worded that the necessary and only reasonable construction to be given disclosure by one skilled in the art is one which will lend clear support to each positive limitation in count.

United States Court of Customs and Patent Appeals,

[Reversed.]

December 14, 1967

Appeal from Patent Office, Interference No. 91,467

Andrew E. Taylor, William R. Hinds (Roberts B. Larson, Larson and Taylor, of counsel) for appellant.

William B. Cridlin, Jr., A. Newton Huff, James T. Corle for appellee.

[Oral argument October 6, 1967 by Mr. Hinds and Mr. Cridlin]

Before WORLEY, Chief Judge, and Judges RICH, SMITH, ALMOND, KIRKPATRICK* WORLEY, Chief Judge, delivered the opinion of the court:

The sole issue raised by Dyer 1 in his appeal from the decision of the Board of Patent Interferences awarding priority of invention to Field 2 is whether, as the board held, Field is entitled under 35 USC 120 to rely upon the 1954 filing date of his parent application to establish priority with respect to the subject matter of the single count:

1. A bulky yarn comprising a separate core yarn and separate excess yarn interwoven back and forth in said core yarn, said excess yarn at a number of random points extending through said core yarn and forming a multitude of elongated loops on the outside of said core yarn, said finished yarn having an increase of denier of greater than 50% over the combined denier of the untreated core and excess yarn. (Emphasis supplied)

Senior District Judge, Eastern District of Pennsylvania, sitting by designation.

1 Dyer is involved on application Serial No. 752,067, filed July 30, 1958, as a continuationin-part of Serial No. 544,886, filed November 4, 1955.

2 Field is involved on application Serial No. 790,789, filed January 30, 1959, as a continuation-in-part of Serial No. 450,075, filed August 16, 1954.

The product is illustrated diagrammatically in Figs. 1 and 2 of Dyer's application.

The record shows that the bulky yarn product defined by the count is made by feeding the "core" yarn into an air jet at a rate slightly faster than it is removed, while feeding the "excess" yarn into the same jet at a rate substantially faster than it is removed. The slight "overfeed" of the "core" yarn, coupled with the turbulence of the air stream in the jet, causes the core filament bundle to open up or "bloom" without forming convolutions, thus permitting the "excess" yarn to penetrate in and out of the "core" yarn as the "excess" yarn filaments form loops as a result of their excessive overfeed. The "core" yarn serves as a stress-bearing member in the final product, whereas the "excess" yarn provides the bulk.

3

The controversy finds its genesis in whether the 1954 Field specification contains a "written description" of a bulky yarn having the structure defined by the count limitations to which we have added emphasis above, viz. whether that application discloses that the product consists of separate "excess" yarn interwoven back and forth in the "core" yarn, extending through it at a number of random points, as well as whether it adequately describes the "manner and process of making" that product.

3 In Swain v. Crittendon, 51 CCPA 1459, 332 F. 2d 820, 141 USPQ 811, we pointed out that, where an applicant contends that he is entitled to rely upon one or more previously filed applications under § 120, the previously filed application (s) must comply with the first paragraph of 35 USC 112, which reads:

The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.

Some background as to the disclosures of the 1959 and 1954 Field applications, with reference to the drawings common to each, will facilitate understanding of the issue involved.

Both Field applications disclose a number of different bulky yarn embodiments and techniques for obtaining them. In a first embodiment common to both applications, "excess" yarn 1 alone is supplied from package 6 to air jet 9 (Fig. 1) at a feed rate substantially greater than that at which it leaves the jet. According to the specification, turbulence in the air jet forms a multitude of elongated loops extending from the central region of the yarn, as illustrated in Fig. 4. The bulked yarn is subsequently twisted and wound on bobbin 17. No separate stress-bearing core yarn is incorporated with the bulked excess yarn.

In a second embodiment common to both applications, a stress-bearing "core" yarn 21 is supplied from package 22 and combined or plied at pigtail 8 with the bulked "excess" yarn after the latter emerges from the jet. The combined yarns are wound on bobbin 17 at the same rate at which the "core" yarn is supplied. The resultant product, illustrated in Fig. 5, is described as having loops "completely about" the body of the "core" yarn.