MAHONEY, Circuit Judge.
The Nashua Manufacturing Company brought suit against Lincoln Stores, Inc., for alleged infringement of patent No. 2,208,533, issued by the United States Patent Office to the Nashua Company July 16, 1940, on an application filed by Robert Amory December 20, 1939. Claims 7, 10, 16 and 19 of the patent are in issue. The patent is hereinafter referred to as the Amory patent or as the Purrey blanket, the commercial name for the blanket manufactured under the Amory patent. The case is
Amory's patent has reference to a household blanket preferably made mainly of rayon fibers but which may be made of other artificial fibers.
The warmth of a blanket varies with the number of air passages in the nap of the blanket. These passages have an insulating effect because they entrap the air thus preventing the cold air from reaching the body under the blanket and preventing the warm air emanating from the body from escaping to the outside. A blanket with a very lofty nap has more air passages and consequently greater warmth than a blanket with a lower nap. The warmth of a blanket with a fairly uniform and homogeneous nap, that is a nap which is equally high throughout the blanket, will be more evenly distributed than the warmth of a blanket with a less uniform nap. In addition, a blanket with a lofty and uniform nap has a more pleasing appearance.
Woolen blankets are generally recognized for their warmth because of their lofty and uniform nap. Of course, the quality of the wool used in a particular blanket will reflect upon the loftiness and uniformity of the nap.
The Purrey blanket is characterized in Amory patent as a blanket with "a lofty and homogeneous, though firmly anchored, stiff and resilient nap, consisting essentially or principally of staple rayon fibre" which has "the advantages of an all wool blanket at a far lower cost". Claim 7 of the patent, which is the most important claim in issue, asserts that the Amory blanket is "a low density blanket composed principally of a filling having a napped wrapping of relatively coarse, soft yarn, essentially of random intermingled staple artificial fibres of an average fibre denier of at least 4.00, said fibres being mostly fibres in excess of 2.75 denier loosely wound with and carried upon a small and hard twisted core thread, with a major portion of the volume of the blanket constituted by said wrapping raised in the form of a lofty, homogeneous nap throughout the area and napped thickness of the blanket on both sides thereof, though firmly anchored to the core thread, and a hard-twisted warp interwoven with said filling."
In the specifications, the wrapping of the filling yarn, i.e., the weft or the woof, is described as "preferably slack twisted and undrafted, though with the fibres randomly intermingled". The wrapping "is lightly laid upon and loosely twisted with or loosely wound around so as to be carried by a small but relatively strong core thread (preferably a spun rayon core) which serves as a reinforcement for the napped rayon exterior of the blanket, in which practically all of what originally was the available or exposed wrapping portion of the filling appears as nap of the finished blanket." The warp upon which the filling threads are woven is a light but strong warp preferably of spun rayon fiber. The specifications further state that the filling wrapping must be of a yarn of an average denier of more than 2.75 and that 4.50 to 5.0 or even more is the preferred average denier. The fibers in the wrapping of the filling "should include a substantial portion of fibers of a length 1½" ± ½" in order to secure an adequate number of ends to provide the desirable nap of this invention". Additional springiness may be obtained by using a small percentage of wool fibers with the rayon fibers. In actual practice the Purrey blanket contains about 20% wool fibers and 80% rayon fibers. The wrapping is "slightly or slackly twisted with or around the core and is subjected to little or no drafting" so that the fibers are not parallelized as they would be if subjected to a drafting process. Thus the fibers are randomly intermingled and the napping operation produces a stiff and resilient nap similar to that of high grade wool blankets. The filling thread is strengthened by a core thread, which is buried in the filling with the warp threads. Thus these threads are not greatly affected by the napping process and the wrapping may be more effectively napped. After the napping operation the
Blankets made essentially of rayon had been known for many years but until the Purrey blanket was placed on the market no rayon blanket achieved any great degree of commercial success. This was due to the fact that the nap of the rayon blankets prior to Amory was not lofty or uniform and shed easily. Men skilled in the art attributed this to the unelastic, straight, smooth qualities of rayon, and thus they believed that rayon could not be used to produce a blanket similar in quality to a wool blanket. Wool, however, is elastic, curly and rough.
The district court found that the nap produced by application of the Amory patent has the same qualities of warmth and much the same appearance as the nap of an all wool blanket of greater cost, and that the Purrey blanket is light, strong, relatively non-shedding, warm and capable of high moisture absorption. It attributed the successful manufacturing result of Amory's process to the use of essentially staple rayon fibers of at least 1" in length and having a denier of at least 4.00 kept in a randomly intermingled state, and to the use of a core thread around which the wrapping is loosely wound, which serves as an anchor for the ends of the fibers so that the nap is held in place and will not shed easily in the napping process or in laundering.
The defendant challenges claim 7 of the Amory patent as not meeting the standards of clarity required by R.S. § 4888, 35 U.S.C.A. § 33, because it is impossible to determine whether a given blanket infringes the patent. This challenge is based on the defendant's assertion that the word "staple" as used in the claim has no exact meaning. The defendant asserts that the word "staple" is used merely to distinguish short filaments from long filaments; that staple rayon is made by cutting long filaments into relatively short lengths or fibers, ordinarily of uniform length; that staple fibers are found in both the uniformly cut fibers and in garnetted material. The district court held, however, that the word "staple" as commonly used in the rayon industry and as used in this patent, means a filament cut to a predetermined length and that this definition read in the light of the specifications requires that the filling wrapping contain fibers of a predetermined length, preferably 1" to 2", and it concluded that there was no difficulty in deciding which blankets infringe and which do not. We agree with the district court.
Patentees are allowed much latitude in terminology and the language they use will be given the meaning intended by them if it can be ascertained from the context. Strong-Scott Mfg. Co. v. Weller, 8 Cir.1940, 112 F.2d 389, 394. It is immaterial that we must refer to the specifications for an understanding that the fibers are to be cut to a length of between 1" to 2". Skinner Bros. Belting Co. v. Oil Well Improvements Co., 10 Cir.1931, 54 F.2d 896, 899.
The defendant further asserts that claim 7 is vague because the phrase "lofty, uniform, homogeneous nap" is meaningless. We agree with the finding of the district court, based on the evidence, that the phrase means a fairly high nap that on cursory inspection does not appear to be appreciably denser in one part than another.
The defendant has cited a number of prior patents, uses and publications, as constituting anticipation of Amory's patent. Although the prior art discloses many and possibly all of the elements used in Amory's invention, no one before Amory had combined all these elements into a single process. It is our understanding Amory did not patent any single element but rather a combination of numerous elements.
The Carleton patent, No. 16870, applied for and granted in 1857, was for the manufacture of cotton flannel. This patent disclosed the use of a "hard warp and filling" for the purpose of allowing the strength necessary to permit a high degree of napping. Amory may have used this principle for the purpose of securing transverse strength but he combined it with many other features not mentioned in the Carleton patent. In addition the Carleton
The Whitman patent, No. 1,249,266, applied for in 1916 and granted in 1917, pertains to the construction of a blanket and disclosed the use of artificial fibers of a uniform length mixed with cotton or wool or flax. The Whitman patent called for spun warp and weft yarns, the construction ordinarily used for wool blankets. Although Whitman did disclose the use of rayon fibers (artificial silk) he recommended the use of waste material to be broken up into short unitary stock and mixed with cotton and spun into a thread. The Purrey blanket differs from the Whitman patent in many respects. Amory does not use waste material, but rather staple rayon cut into short fibers of a predetermined uniform length. He does not spin the rayon fibers into a thread, but leaves them undrafted and randomly intermingled. Whitman did not use the principle of a cored yarn as did Amory and Carleton. Whitman's patent did not specify the denier of the rayon filaments, but required "short fine fibers" as distinguished from the fibers of relatively coarse denier, specified by Amory. There is no evidence that the filling yarn of the Whitman patent could be napped so as to produce the lofty, uniform nap of the Purrey blanket, nor is there any evidence that the Whitman blanket was ever made for commercial purposes or otherwise.
The Cobb patent, No. 2,012,184, applied for in 1934, granted in 1935, particularly for the manufacture of blankets, disclosed the use of a cored filling weft yarn comprised of a closely twisted relatively small core wound around and with a loosely spun cotton or other fiber thread to provide the transverse strength necessary to permit the wrapping to be effectively napped. But Cobb did not combine this element of Amory's patent with the many other constituents used by Amory.
The Sowter patent, No. 2,044,130, applied for in 1934, granted in 1936, shows the use of a core and wrapping made of artificial fibers, but nothing more.
The filling thread of the Neaves patent, No. 2,090,547 (applied for in 1936, granted in 1937) is composed of a core of cotton or artificial fiber and a wrapping of lightly spun wool yarn in an undrafted state. The construction used by Neaves is very similar to the construction used by Amory. But the Neaves patent did not use artificial fibers in the wrapping. Neaves used wool. Amory used rayon. The defendant asserts, however, that the mere substitution of rayon for wool does not constitute invention. Ordinarily the mere substitution of a known equivalent for one of the elements of an old structure does not constitute invention. Crouch v. Roemer, 1880, 103 U.S. 797, 799, 26 L.Ed. 426; Smith v. Nichols, 1874, 88 U.S. 112, 119, 21 Wall. 112, 119, 22 L.Ed. 566. But this is not always true. Smith v. Goodyear Dental Vulcanite Co., 1876, 93 U.S. 486, 492, 23 L.Ed. 952. The attainment of the result reached by Amory required more than merely substituting rayon fibers for the wool fibers used by Neaves. The qualities of the two materials are vastly different. What would have resulted from a mere haphazard substitution is highly speculative. For years, men skilled in the art had endeavored to substitute rayon for wool in the manufacture of blankets with little success. That made it amply clear that such a substitution would not achieve the desired result without knowledge of the particular type of rayon fiber needed. Amory was the first to disclose that the fibers must be of relatively coarse denier and that they should be cut to a specific predetermined length. Without the disclosures made by Amory, we cannot say that a substitution would have been successful. Nor can we say that the substitution specified by Amory was apparent to a man skilled in the art. The many years of futile research and labor by men skilled in the art is weighty evidence of this. Williams Iron Works Co. v. Hughes Tool Co., 10 Cir. 1940, 109 F.2d 500, 510.
We agree with the district court that the prior patents do not anticipate Amory. Carleton, Cobb, Sowter and Neaves all used a cored filling. Whitman used rayon fibers of a predetermined length. Neaves left his fibers in an undrafted
The defendant also cites certain publications prior to Amory as involving anticipation. We do not believe that these publications do any more than disclose certain particular elements utilized in Amory's combination. Their function is no greater in this case than the function of the prior patents already discussed. These publications discuss the demand for short rayon staple fibers and their superiority to garnetted rayon, but recommend that, after the filaments are cut to the desired length, they be spun into a thread. One of these publications defines rayon staple as a primary product having standard length and denier and rayon waste as a secondary product, somewhat irregular as to uniformity. The Rayon and Synthetic Yarn Handbook, Schwartz & Mauersberger, 1936, states that the most popular length of rayon staple fiber is 1½ inches, and that staple fiber is made in three degrees of fineness: 1.5 denier, 3 denier and 4.5 denier. This handbook also discloses that "5 denier or higher filaments are commonly employed in the making of coarse yarns and also so-called `artificial wool,' which is staple rayon fiber cut to lengths approaching the length of the wool fiber with which it is to be mixed." There is evidence that the ordinary staple length of wool fibers varies between one and two inches and averages about 1½ inches. And there can be no doubt that staple rayon fibers cut to a predetermined length of one to two inches and having a denier of 4.00 or higher were known to the art prior to Amory. But these publications do not teach the lesson which the art awaited. They only teach the existence of the materials utilized by Amory. They do not disclose, as did Amory, that when rayon fibers are used in a blanket construction they must be 1" - 2" in length and of a denier of at least 4.00 and kept in a randomly intermingled state, loosely wound around a core thread. Amory's precise combination was unknown to the art, although, perhaps, all of its elements had been previously disclosed and utilized. There is no indication from the prior publications or patents that the combination of elements used by Amory would produce a nap having the lofty, uniform and warmth-giving characteristics of the Purrey blanket. The fact that these elements were known for some time prior to Amory and that the successful combination remained undiscovered before Amory's patent is entitled to considerable weight. Rayon staple was known for many years before Amory, but the art had taught that the most beneficial use of these fibers would be derived by spinning them into a thread rather than leaving them undrafted and randomly intermingled. The previous adoption of an essential element of a patent
The defendant further urges that certain prior uses by Marshall Field & Co. and its subsidiaries anticipate the Amory patent. The first of these prior uses is a blanket manufactured between 1927 and 1929 by Carolina Cotton & Woolen Mills Company, formerly a subsidiary of Marshall Field. The filling of these blankets was composed of a cotton core yarn and a wrapping of 92% garnetted rayon waste and 8% wool. The individual rayon fibers varied in length from nothing to 2" or 3". There is evidence that the average length of the fibers was 1.75" before napping and .5 inches after napping and that the denier was 8.0 both before and after napping. It may be that the fibers were randomly intermingled and that the wrapping was loosely twisted around the core. But it is inescapable that these blankets were not of the quality of the Purrey blanket. The nap shed considerably in washing. These blankets experienced little commercial success. Only 2395 of them were manufactured between 1927 and 1929. The import of this prior use in the case at bar is at most no greater than that of the Neaves' patent except that the Carolina Company used garnetted rayon while Neaves used wool. The use of garnetted rayon did not produce the desired result. It was not until Amory used staple rayon of specified length and denier that the result toward which the efforts of the art had been directed was obtained.
The 1932-1933 prior use of the Carolina subsidiary of Marshall Field was essentially the same as the 1927-1929 blanket except that the wrapping of the filling was made up solely of garnetted rayon. Very few, if any, of these blankets were sold.
Between 1934 and 1936, Marshall Field manufactured about 57,000 rayon and cotton blended blankets. This prior use is not alleged to be anticipation and adds nothing to the prior uses already discussed.
In 1938, Marshall Field developed a blanket containing 50% rayon staple and 50% cotton. But as late as 1938 Marshall Field did not know Amory's secret. The blanket was made in accordance with the construction of the cotton system. The filling contained no core; the fibers were highly drafted and spun into a thread.
The defendant urges that the commercial failure of the prior uses of Marshall Field and its subsidiaries cannot be used to undermine the weight to be accorded them in the determination of the question of anticipation. It asserts that the absence of commercial success was due to the poor quality of rayon from 1927 to 1938, and that the quality of rayon has constantly improved since 1927, so that a commercially successful blanket containing principally rayon fibers is now feasible; and that the price of rayon until recently was so high that a rayon blanket could not compete in the commercial market with an all-wool blanket. We believe that the evidence supports the findings of the district court which were adverse to the defendant on both contentions. The difference in quality and function between staple rayon and high grades of garnetted rayon is one of the very reasons for the success of the Purrey blanket. The large number of very short fibers contained in garnetted rayon prevents the production of a comparatively permanent lofty, uniform nap. In washing, and even in the napping operation, the shorter fibers are pulled out, thus destroying the lofty and uniform qualities necessary for the achievement of a blanket which will afford warmth over a long period.
On the price question, the defendant cites the differential between the prices of rayon and cotton. The analogy to be drawn must be between the prices of rayon and wool, for the purpose of the Amory patent was to create a rayon blanket that could compete with a high-grade wool blanket. Except for a brief period in 1932, the price of wool per pound was almost twice as much (and sometimes much more) as the price of rayon staple per pound between 1928 and 1939. During that period the price of rayon was an incentive rather than a deterrent for those who sought to produce a rayon blanket that could successfully compete with a woolen blanket.
From October 1, 1939 to early 1945, 2,500,000 blankets were sold under the
We come now to the question of whether the combination of elements utilized by Amory constitutes a patentable invention. A mere aggregation of a number of elements known to the prior art, each of which performs the same function for which it was previously utilized and which produces no new result or no new cooperative function, does not constitute a patentable invention. Lincoln Engineering Co. of Illinois v. Stewart-Warner Corp., 1938, 303 U.S. 545, 549, 58 S.Ct. 662, 82 L.Ed. 1008; Grinnell Washing Machine Co. v. E. E. Johnson Co., 1918, 247 U.S. 426, 433, 38 S.Ct. 547, 62 L.Ed. 1196; Computing Scale Co. of America v. Automatic Scale Co., 1907, 204 U.S. 609, 616, 27 S.Ct. 307, 51 L.Ed. 645. And the addition of a new and useful element to an old combination is not a patentable invention if it results from no more than the exercise of the mere skill of the calling and is plainly indicated by the prior art. Electric Cable Joint Co. v. Brooklyn Edison Co., 1934, 292 U.S. 69, 79, 54 S.Ct. 586, 78 L.Ed. 1131. Nor is the mere substitution of one material for another or the judicious selection of materials which produces an expected result and does not involve a change of method or develop a novel use a patentable invention. Florsheim v. Schilling, 1890, 137 U.S. 64, 76, 11 S.Ct. 20, 34 L.Ed. 574; Kalich v. Paterson Pacific Parchment Co., 9 Cir. 1943, 137 F.2d 649, 651. But the fact that a new combination of old elements produces a new and beneficial result, hitherto unattained, merits consideration as evidence of invention. Webster Loom Co. v. Higgins, 1881, 105 U.S. 580, 591, 26 L.Ed. 1177. And a substitution which involves a new mode of construction, or develops new uses and properties of the article formed, may amount to invention. Smith v. Goodyear Dental Vulcanite Co., supra, 93 U.S. at page 492, 23 L.Ed. 952, distinguishing Hotchkiss v. Greenwood, 11 How. 248, 13 L.Ed. 683. A patentable combination may be comprised of elements that are all old or all new, or partly old and partly new. Leeds & Catlin Co. v. Victor Talking Machine Co., (No. 2), 1909, 213 U.S. 325, 332, 29 S.Ct. 503, 53 L.Ed. 816. Where none of the patents in the prior art disclose all the elements of the patent in issue, they cannot be combined for that purpose, Cover v. Chicago Eye Shield Co., 7 Cir.1940, 111 F.2d 854, 859. An inventor cannot be denied the rights derived from a patent merely because he has combined elements, all of which were known to the prior art, if he has produced something involving novelty and utility. National Aluminate Corp. v. Permutit Co., 8 Cir. 1944, 145 F.2d 175, 179; Monarch Marking System Co. v. Dennison Mfg. Co., 6 Cir., 1937, 92 F.2d 90, 94. A finding of invention may be warranted where the patentee combines features, even though all of them are known to the prior art, which produce a new result, especially where that combination enjoys marked commercial success. Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 1944, 321 U.S. 275, 279, 64 S.Ct. 593, 88 L.Ed. 721; Halliburton Oil Well Cementing Co. v. Walker, 9 Cir.1944, 146 F.2d 817, 819; Texas Rubber & Specialty Corp. v. D. & M. Machine Works, 5 Cir.1936, 81 F.2d 206, 208. Amory's combination produced a new result, although it may have been composed
The defendant charges that Amory's combination was simple, obvious to anyone skilled in the art. That on retrospect the device appears simple or obvious does not necessarily establish a lack of invention. Webster Loom Co. v. Higgins, supra, 105 U.S. at page 591, 26 L.Ed. 1177. It is indeed significant in this respect that the workers in the art had labored long and extensively to produce the blanket first produced by Amory. Potts & Co. v. Creager, 1895, 155 U.S. 597, 608, 15 S.Ct. 194, 39 L.Ed. 275. Retrospection is often deceptive and cannot be accorded recognition in the law pertaining to patents. Diamond Rubber Co. v. Consolidated Rubber Tire Co., 1911, 220 U.S. 428, 435, 31 S.Ct. 444, 55 L.Ed. 527. At least Amory provided the final step that proved the difference between success and failure. This is a factor which has been accorded considerable recognition in the courts. The Barbed Wire Patent, 1892, 143 U.S. 275, 283, 12 S.Ct. 443, 36 L.Ed. 154; Consolidated Safety-Valve Co. v. Crosby Steam Gauge & Valve Co., 1885, 113 U.S. 157, 179, 5 S.Ct. 513, 28 L.Ed. 939.
Whether the question of invention is one of fact or of law is not too clear on the authorities; this uncertainty is indeed not entirely dispelled by consideration of decisions of the Supreme Court. See the discussion by Judge Woodbury in Hanovia Chemical & Mfg. Co. v. David Buttrick Co., 1 Cir. 1942, 127 F.2d 888. In Thomson Spot Welder Co. v. Ford Motor Co., 1924, 265 U.S. 445, 446, 44 S.Ct. 533, 534, 68 L.Ed. 1098, the court said: "The question whether an improvement requires mere mechanical skill or the exercise of the faculty of invention, is one of fact; and in an action at law for infringement is to be left to the determination of the jury." And in the recent case of Goodyear Tire & Rubber Co., Inc. v. Ray-O-Vac Co., 1944, 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721, the majority opinion certainly treated the question as one of fact. Until advised to the contrary by the Supreme Court, we shall so regard it. The presumption of the validity of Amory's patent arising from its issuance, Mumm v. Jacob E. Decker & Sons, 1937, 301 U.S. 168, 171, 57 S.Ct. 675, 81 L.Ed. 983, taken together with the many years of effort to produce what Amory produced and its substantial commercial success, makes it impossible for us to say that the district court was clearly erroneous in deciding that Amory had produced a patentable invention. Rule 52(a), F.R.C.P., 28 U.S.C.A. following section 723c.
We shall consider briefly the other claims in issue held to be valid and infringed by the court below. Claim 10 adds to claim 7 the statement that the Amory blanket has a moisture absorption factor of more than 10%. This may well be an important function of the Amory blanket, but it is nevertheless no more than a function performed by the coaction of the various ingredients of the blanket. The claims of a patent cannot be broadened by a description of its functions or the results produced by its elements. United Carbon Co. v. Binney & Swift Co., 1942, 317 U.S. 228, 234, 63 S.Ct. 165, 87 L.Ed. 232; Rice v. Nash-Kelvinator Corp., 6 Cir.1945, 150 F.2d 457, 460. The allowance of a claim based on results inherent in the use of the construction described would extend the patent monopoly granted "beyond the discovery, and would discourage rather than promote invention." Holland Furniture Co. v. Perkins Glue Co., 1928, 277 U.S. 245, 256, 257, 48 S.Ct. 474, 479, 72 L.Ed. 868. Hence, claim 10 is invalid insofar as it purports to claim more than is claimed by claim 7.
Claim 16 differs from claim 7 in that it specifies that the wrapping shall be "drafted less than 2.00" and that the fiber denier shall be at least 2.75, instead of at least 4.00. Although this claim adds little to claim 7, it contains different, specific information as to the amount of drafting and the denier of the fibers and as such constitutes a distinct and separate claim. Claim 16 is sustained on the reasoning set forth in the discussion of claim 7.
The question of infringement has not been argued before us, but it is quite clear that the defendant's Jacquard blanket does infringe.
The judgment of the District Court is affirmed as to claims 7 and 16 and reversed as to claims 10 and 19 of patent No. 2,208,533, and the case is remanded to that court for further proceedings not inconsistent with this opinion.
WOODBURY, Circuit Judge (dissenting).
I am willing to agree with my associates in regarding the question of invention as one of fact until advised to the contrary by the Supreme Court. I would, however, like to point out that although the question is treated as one of fact in the majority opinion in Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 64 S.Ct. 593, 88 L.Ed. 721, decided in 1944, three of the dissenting Justices in that case indicated in a footnote appended to their opinion that on the authority of Mahn v. Harwood, 112 U.S. 354, 358, 5 S.Ct. 174, 6 S.Ct. 451, 28 L. Ed. 665, decided in 1884, and cases collected in United States Supreme Court Digest (West 1943), Patents, Vol. 11, they viewed the question as one of law.
Furthermore, I agree with my associates' analysis of Amory's contribution. That is to say, seeing nothing new in the warp threads which he used, in the way those threads were woven with his filling threads, in a filling thread consisting of a hard twisted core wrapped with slack twisted and undrafted artificial fibers, or in his napping process, I conclude that Amory's contribution was the use as wrapping for his filling threads of a large proportion of relatively coarse artificial fibers a substantial portion of which were uniformly 1½" ± ½" long, i.e., staple rayon, a known material, whereas the prior art had used for that purpose artificial fibers of short but random lengths and denier, i.e., garnetted rayon, another known material. Thus I conclude, as my associates do, that Amory's contribution consisted in the substitution of one known material for another in an old combination. My disagreement in this case stems from my inability under these circumstances to reconcile the finding of invention made below with the rule of law enunciated by the Supreme Court in 1850 in the case of Hotchkiss v. Greenwood, 11 How. 248, 13 L. Ed. 683, since in my view Amory's substitution of materials did not produce a rayon blanket differing in kind or species from those which had preceded it, (Smith v. Goodyear Dental Vulcanite Co., 93 U.S. 486, 23 L.Ed. 952, decided in 1876, see also our opinion on rehearing in Grant Paper Box Co. v. Russell Box Co., 154 F.2d 729); but only produced a better rayon blanket than had previously been known.
The rule established by the Hotchkiss and Smith cases, supra, which over the years since its enunciation has been many times applied, is that it does not amount to invention to substitute one known material for another in an old combination when the new product resulting differs from the old only in that it possesses desirable qualities or characteristics of the old in an accentuated form or to a greater degree, but that such substitution may amount to invention if the new product is an improvement over the old for the reason that it differs therefrom in kind or species having new uses, functions or properties. Thus in the Grant Paper Box Co. case supra, we were able to find invention because the patentee's laminated paper not only differed in degree from the laminated papers of the prior art in that it was thinner and more highly moisture-vapor proof, but also differed from them in kind in that it could be flexed without cracking and could be written or printed upon, — qualities not possessed in any appreciable
Hence at the threshold of cases of this sort a distinction must be drawn between differences in degree and differences in kind, and only when a difference falling in the latter category is found to exist does the question of invention arise. From this it follows that we do not come to the question of patentable invention in the case at bar until we are able to discover something in addition to a difference in degree between the Amory blanket and the rayon blankets of the prior art.
Obviously differences in degree cannot be distinguished from differences in kind with anything even remotely approaching mathematical nicety. For this reason the distinction is unsatisfactory in that each case must be decided on its own facts as it arises and in close cases like the present differences of opinion are normally to be expected. As I see it, the Amory blanket differs from the rayon blankets which preceded it only in that it is better in appearance, sheds less in laundering, and is not only warmer, but is also more evenly warm throughout, because it possesses a loftier and more homogeneous nap. Hence I conclude that Amory's substitution of materials produced a rayon blanket differing from earlier ones in degree of excellence only — that Amory's substitution only accentuated desirable qualities possessed to less degree by the rayon blankets of the prior art — and thus that his patent is invalid under the rule of Hotchkiss v. Greenwood, supra.