Appellants filed an application titled "Nickel Chromium Base Alloy Products," serial No. 855,076 on November 24, 1959,1 and have appealed from an adverse decision on patentability with respect to claims 1-10 therein. The Board of Appeals reversed the examiner's holding of unpatentability over the art of record as to the alloy defined by claims 11 and 12.

The invention is directed to articles of alloy compositions of which the main components are chromium, cobalt, and nickel, with minor percentages of manganese and silicon, and three elements which function as hardeners, aluminum, titanium and molybdenum.

Alloy products of the compositions are used in the production of the inner walls of "jet [exhaust] pipes," and possess two advantageous welding characteristics, (1) ability to withstand rapid, repeated, thermal cycling (thermal shock resistance), and (2) a relatively high ductility.

The three hardeners are each limited to a specific range of composition, and additionally the sum of 1x percent of Mo + 2x percent of A1 + 4x percent of Ti must be "below approximately 16." That value is called the "balance factor." The aluminum content is disclosed to be of considerable importance. If it exceeds .74%, "ductility of the alloy falls and its welding properties are adversely affected due to skin formation." If the aluminum content is "substantially less than .3%[,] the resistance * * * to deformation under stress falls off to a low value."

"1. A formed metal article having a weld therein, the metal of said article and of said weld consisting by weight essentially of approximately 19% to 23% chromium, approximately 12% to 25% cobalt, each of the three hardening elements molybdenum, aluminum and titanium within the approximate ranges of 3.0% to 8.6% molybdenum, 0.3% to 0.74% aluminum, 1.7% to 2.45% titanium, together with approximately 0.2% to 0.6% manganese, and 0.1% to 0.5% silicon, and the balance nickel apart from impurities and residuals from de-oxidizers, said alloy being characterized in that the sum of the percentage of molybdenum plus twice the percentage of aluminum plus four times the percentage of titanium equals a figure below approximately 16."

It is to be noted that applicants are claiming an article of manufacture which contains a weld. In certain claims both the weld and the article are composed of the alloy and in others only the weld is so specified. The remaining claims are drawn to preferred ranges of the three hardener elements.

The reference relied on for the rejection is:

Gresham et al. (hereinafter Gresham) show the same basic alloy of Cr-Co-Ni and the additives Mn and Si, all in the same proportions as claimed by appellants. Gresham also is particularly concerned with the same three hardeners, Mo, A1 and Ti, but the balance factor is selected to be between 16 and 20 for the purpose of safe control of the creep strength in blades and other "high duty" engineering parts of gas turbines.

Reference to the accompanying graph will aid in understanding the issues here:

NOTE: OPINION CONTAINING TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The graph2 represents a plot of the amount of two of the hardeners, aluminum vs. titanium. It should be understood that the amount of the third hardener, molybdenum, is kept within a defined range. Since that range is the same for both appellants and Gresham, we shall follow appellants' lead in not according it further significance.

Referring to the graph, appellants' claims are drawn to articles, in part consisting of alloys, falling within the area ABCD, and, as preferred alloys, within the area GHIJ. Appellants' specification discloses three specific examples of alloy compositions which fall within claimed area ABCD, positions labeled Ex. 1, Ex. 2, and "X" on the graph. The alloys Ex. 1 and Ex. 2 have a balance factor below 16 while that of alloy "X" is above 16.3 The specification also discloses thermal shock tests of these three compositions. A weld of the alloys of Ex. 1 and 2 withstand 1410 and 1020 thermal cycles before breaking, while that of alloy "X" failed after 510 cycles. The superiority of the two alloys having a balance factor below 16 seems clear.

Two alloys of Gresham having balance factors below 16, termed "f" and "g", are located on the graph at positions R"f" and R"g" respectively. In two affidavits by Meetham, and an exhibit, appellants attempted to duplicate the Gresham alloys "f" and "g". The duplications resulted in alloys of composition labeled A1 and A2 on the graph. These affidavit alloys, in comparison with the thermal shock characteristics of alloys Ex. 1 and 2, failed after 488 and 64 cycles respectively, although both alloys A1 and A2 had balance factors below 16. Those results clearly show that the amounts of the hardeners, particularly aluminum, are, as the specification states, "of considerable importance." While appellants' attempt to duplicate alloy "f" resulted in a somewhat poorer approximation than the duplication of "g", the reasonable approximation of such tight specifications is entirely acceptable and clearly shows the criticality of the aluminum component.

While alloys "f" and "g" fall outside appellants' claimed area ABCD, the examiner relied on other disclosure in Gresham as showing that "the alloys [per se] are old." Appellants acknowledge only that Gresham discloses some 12 other alloys, of which 5 are said by Gresham to have balance factors of less than 16, and which fall somewhere within the area WXYZ of the graph above. Appellants argue that Gresham does not describe any specific alloy falling within area ABCD and having a balance factor less than 16, and thus does not place the disclosed alloy in the possession of the public, In re LeGrice, 301 F.2d 929, 49 CCPA 1124; In re Brown, 329 F.2d 1006, 51 CCPA 1254; E. I. du Pont de Nemours & Co. v. Ladd, 117 U.S.App. D.C. 246, 328 F.2d 547 (1964).

That part of Gresham relied on by the examiner as containing anticipating alloys states:

"* * * In making a considerable number of casts from alloys of the type containing, besides 20% Cr and 20% Co, 6 per cent Mo, 0.9 per cent A1 and 2.4 per cent Ti, the balance being Ni, in varying the percentage of Mo by ± 0.4 per cent, of A1 by ± 0.3 per cent and of Ti ± 0.3 per cent, we found that, if all errors were towards the minus side and the balance factor dropped to 15.2, the creep strength was poor. On the other hand, whenever one of the three elements kept on the higher level, the creep properties were greatly improved.

"A series of twelve nickel alloys were prepared having the approximate composition 20% chromium and 20% cobalt, while the percentage of Mo ranged from 5.5 to 6.5, that of A1 from 0.4 to 1.2, that of Ti from 2.2 to 2.7. * * *" [Emphasis supplied.]

Gresham continues by describing a graph of the time to fracture plotted against the balance factor for the twelve alloys. The graph contains twelve data points, five of which correspond to alloys with balance factors between 15 and 16.

The solicitor takes direct issue with appellants' contention that Gresham does not show the claimed alloys to be old. The solicitor first points to the abovequoted portion of Gresham and notes that among the alloys there is one "in which all variations are on the minus side and the balance factor is 15.2 That alloy obviously contains 5.6% Mo, 0.6% A1 and 2.1% Ti." Such an alloy would fall at the intersection of lines E-K and G-H on the graph reproduced above, and has been labeled O.

The solicitor then finds a second alloy to fall specifically within appellants' claimed area:

"* * * Second, of the twelve alloys described by Gresham et al. in column 2, lines 36-64, one is shown on their graph * * * by an (x) at a point corresponding to an alloy with a balance factor of 15.1. As shown by the balance factor formula (1 X % Mo + 2 X % A1 + 4 X % Ti=balance factor), there is one, and only one, alloy within the ranges of 5.5-6.5% Mo, 0.4-1.2% A1 and 2.2-2.7% Ti which has a balance factor of 15.1. That alloy is the one containing the minimum of 5.5%Mo, 0.4% A1 and 2.2% Ti. On figure 2 of appellant's drawings * * *, 0.4% A1 and 2.2% Ti are at a point which is within the parallelogram GHIJ. On the graph * * * [herein], 0.4% A1 and 2.2% Ti are at the point Z."

We find no error in the solicitor's analysis and hold that Gresham does disclose that two alloys, alloys "O" and "Z" which fall within the terms of the claims, were indeed prepared. There is thus no question that the test in In re LeGrice, supra, and In re Brown, supra, have been met.

Appellants further contend that even if the alloys claimed are old, the article of manufacture would not have been obvious, since Gresham was interested in the above 16 balance factor alloys for another purpose, and since the ductility and thermal shock properties employed by appellant were not disclosed by Gresham.

We do not agree with that contention. We recognize that the claims do not call for the alloy per se, being directed to articles which in part contain a weld of the specified alloy composition. Finding the composition fully disclosed in Gresham, we do not think that the remaining claim limitation, of an article and weld, to be sufficient to impart patentability to the claims, even if we take the claims as drawn only to a new use. We agree with the examiner that it is known in the art to weld alloys. Further, we agree with the examiner that one of ordinary skill in this art would know that alloys having poor creep strength, clearly and repeatedly disclosed by Gresham as a property of those alloys having a balance factor below 16, correspondingly would have a higher ductility.4 We find nothing unobvious in employing the known alloy to take advantage of one of its properties, poor creep strength, or the corresponding property higher ductility.

Appellants rely on In re Tanczyn, 202 F.2d 785, 787, 40 CCPA 886, 889, wherein the claims were directed:

"* * * to an article of manufacture produced from an alloy composition covering a particular range, with the prior art showing two specific alloy compositions falling within the range recited in the claims. * * *

* * * * * *

"* * * the appealed claims are not directed to an alloy composition as such, but to certain manufactured products. Specifically, the claims are directed to `wrought and polished straight chronium stainless steel products * * * substantially free of surface defacing complex silicate inclusions * * *.' In these particular circumstances, we think such limitations are significant and give life and meaning to the appealed claims. It was the search for such products by appellant which gave rise to discovery of the cause of the problem here involved, and to its solution. Further, we agree with the argument by counsel for appellant that the distinction between claims to the wrought and polished products and any claim to the alloy composition as such is real and significant, since production of the alloy composition itself would not infringe the appealed claims to the wrought and polished products." [Emphasis by the court.]

In distinction to the facts of the Tanczyn case, we do not find in the particular circumstances here that the claimed limitations "are significant and give life and meaning to the appealed claims." To the contrary, it is precisely our point that it is the alloy composition here which gives meaning to the instant claims. Further, we find in Tanczyn a recitation of a product not known or produced before as resulting from the discovery of the cause of a problem and the discovery that the solution lay in employment of the particular alloys. We have no such situation here.

We find In re Petering, 301 F.2d 676, 49 CCPA 993; Traitel Marble Co. v. Hungerford Brass & Copper Co., 18 F.2d 66 (2d Cir. 1927), Tilghman v. Proctor, 102 U.S. 707, 26 L. Ed. 279 (1880), and International Nickel Co. v. Ford Motor Co., 166 F. Supp. 551, (S.D. N.Y.1958) inapposite to the facts of this case. Here, the Gresham disclosure is an adequate description of the alloys within the meaning of 35 U.S.C. § 102, as distinct from certain reference compounds in Petering. In contrast to the Traitel Marble, Tilghman, and International Nickel cases, there is in Gresham more than an unrecognized or unwitting production of the alloys now claimed, since the alloys were described as actually produced pursuant to a specific purpose and were tested in a "very sensitive creep test." The selection by Gresham of only those alloys which suited his immediate purpose does not transform the disclosure of the other specific below-16 alloys or their properties to a state of such obscurity as to make their selection for another use, by employing their clearly desirable properties, unobvious.

For the foregoing reasons the decision of the board is affirmed.

Affirmed.

SMITH, Judge (dissenting).

The issue in this case is hazy and shifting and the confusion begins with the examiner's rejection of the appealed claims as "unpatentable over" the Gresham et al. U. S. patent No. 2,712,498, issued July 5, 1955. An indication of the range of rejections which the examiner included within the ambit of this stated ground of rejection is found in the summary of the examiner's answer from which we learn that the examiner's position is:

(1) The alloys are old.

(2) It is known in the art to weld such alloy.

(3) Such alloys are commonly used in high-strength, high-temperature service.

(4) There is an insufficiency of test results to support either criticality of the claimed ranges or a patentable distinction over Gresham et al.

(5) It would be obvious to select those alloys of Gresham et al. with the lower creep strengths for their inherent complimentary properties of higher ductility and thermal shock resistance.

(6) There was no original disclosure of criticality of Mn and Si.

(7) The alloys differ in degree only, rather than in kind from those of Gresham et al. with balance factors above 16.

Except as it affirms the rejection of the appealed claims, the action of the board remains something of a mystery. For example, in the final rejection of September 11, 1961, the examiner stated "It is agreed that technically speaking applicants' claimed alloys embrace novelty over alloys (f) and (g) of patentees." This position changes somewhat in the examiner's answer which states:

"* * * While the specific alloys (f) and (g) were slightly outside the now claimed range of appellants for Al, it nevertheless must be remembered that the examples are meant to be representative of the alloy ranges. `References are valid for what they convey, explicitly or implicitly and therefore the mere fact that experimentation on the old alloy did not, from the patent of the old alloy, appear promising is not important,' In re Aller [220 F.2d 454, 42 C.C.P.A. 824] 105 U.S.P.Q. (BNA) 233. * * *"

However, by the time this aspect of the case is treated by the board we find the alloys (f) and (g) of Gresham et al. are relied upon, as stated by the board, to "show amounts for these metals which fall within the claimed ranges."

"In the final analysis, we find that the claimed alloys are old since they encompass alloys (f) and (g) of Gresham et al. with only a slight difference in the aluminum content, i. e., the difference between .74% (appellants' upper limit) and .85% (alloy (g)) and .87% (alloy (f)), which, incidentally, are only representative of the alloy ranges disclosed in the reference.

"For the above reasons, we are, therefore, in agreement with the Examiner that the claimed alloys are old even though they have been rejected by Gresham et al. for their particular purpose, viz., high creep resistance. * * *"

The opinion of the board concludes with reversing the examiner as to the rejection of claims 11 and 12 and affirming the rejection of presently appealed claims 1-10.

In attempting to articulate the grounds for the board's affirmance, appellants in their brief have, I think, correctly summarized the problem as follows:

"The Board affirmed the Examiner's rejection of claims 1-10 on grounds which, while not stated to be new grounds, show that the Board adopted a materially different reading and appraisal of the Reference (R-45-7). Its affirmance was rested on two grounds neither of which had been relied on in the Final Rejection — (1) lack of evidence of criticality as to the 0.74% upper limit for aluminum (R-45-6), this being the ground the Examiner had withdrawn after the first Meetham affidavit was filed, and (2) substantial anticipation by alloys (f) and (g) of the Reference (R-46-7), this being the ground the Examiner had admitted in the Final Rejection was not `technically speaking' available."

At best the rejection which the majority affirms is a "hydra-headed monster" which may be either a rejection under 35 U.S.C. § 103, with all its many problems and implications, or a rejection under section 102(b), with its own manifold problems. One result of this confusion is a majority opinion here, which I must confess, I simply do not understand, except that it affirms something which happened below; and that "something," translated into language which I hope appellants may understand, is that they are denied claims 1-10. I share what I am certain will be their consternation in trying to ascertain from the majority opinion the statutory basis upon which the result is predicated.

For example, after finding In re Petering, 301 F.2d 676, 49 CCPA 993; Traitel Marble Co. v. Hungerford Brass & Copper Co., 18 F.2d 66 (2d Cir. 1927); Tilghman v. Proctor, 102 U.S. 707, 26 L. Ed. 279 (1880); and International Nickel Co. v. Ford Motor Co., 166 F. Supp. 551, (S.D.N.Y.1958) "inapposite to the facts in this case," the majority states:

"* * * Here, the Gresham disclosure is an adequate description of the alloys within the meaning of 35 U.S.C. § 102, as distinct from certain reference compounds in Petering. * * *"

Yet, this does not appear to be the basis for the majority's decision for we find the concluding sentence of the same paragraph states:

"* * * The selection by Gresham of only those alloys which suited his immediate purpose does not transform the disclosure of the other specific below-16 alloys or their properties to a state of such obscurity as to make their selection for another use, by employing their clearly desirable properties, unobvious."

Thus in the same paragraph of the majority opinion we find the affirmance predicated on both sections 102 and 103. Lest anyone think my preoccupation with the foregoing matters is unduly technical, I point out again that such a result does violence to both the spirit and the letter of 35 U.S.C. § 132. Does the shifting, changing manner of rejection here advance the real public interest recognized in Article I, Sec. 8 of the Constitution? Does it reflect an appreciation for or any attempt to implement the Congressional intent underlying sections 102, 103 and 132 of the Patent Act of 1952? I think it does not.

Passing now to the merits of the case, I disagree that the invention claimed in appealed claims 1-10 is "described" in the Gresham et al. patent as required under section 102 or that it was "obvious" in view of Gresham as required under section 103.

The major premise upon which the rejection is based is that "the alloys are old." I do not agree that the claimed alloys were "old" in Gresham et al. in the sense that 102 requires, i. e., that they be "described." Neither do I agree that they are "old" in the sense of being "obvious."

My thinking and my departure from the analysis of the majority begin with an understanding of what is meant by the term "alloy." Metals Handbook (1948 ed.), published by the American Society for Metals, defines "alloy" as "A substance that has metallic properties and is composed of two or more chemical elements of which at least one is a metal." The Condensed Chemical Dictionary (6th ed. 1961) indicates that an alloy is:

"A solid or liquid mixture of two or more metals; or of one or more metals with certain nonmetallic elements by fusing the components. The properties of an alloy are often greatly different from those of the component metals, making them more satisfactory for many uses than any pure metals. * * *"

Van Nostrand's Scientific Encyclopedia (3d ed. 1958), at 1051, under the heading "Metals and Alloys," states:

"It is a well-known fact that many important alloy combinations have properties which are not easy to predict on the basis of the properties of the constituent metals. * * In some cases very small amounts of an alloying element produce remarkable changes in properties * * *.

* * * Other properties which can be developed to a much higher degree in alloys than in pure metals include corrosion-resistance, oxidation-resistance, at elevated temperatures, abrasion-or wear-resistance, good bearing characteristics, creep strength at elevated temperatures, and impact toughness. * * *" [Emphasis added.]

Thus the basic concept of an alloy is that it is an integral entity and, as such, is something more than the sum of its individual constituent components. The opening paragraph of the specification before us states:

"This invention relates to chromium-nickel base alloy products having advantageous welding characteristics. Specifically, such advantageous characteristics include the ability to withstand repeated, rapid, thermal cycling. Products embodying the invention also possess a relatively high ductility." It seems to me it must be admitted that an alloy having such properties is a different entity from anything disclosed in Gresham et al. True, it is made up of the same constituent elements which Gresham et al. use in making their disclosed alloys. Here, however, I think the resemblance ends. The appealed claims, as they define the alloy from which "the article" or the "weld" is formed, specify precise and limited proportions of certain of the constituents which go to make up the new alloy.

Just as wheels, levers and gears are individually old constituents of combinations containing them, so are alumium, molybdenum, titanium and the other constituents of appellants' alloy. Yet in neither case is the new entity which is born from a particular combination of them "described" by a reference showing them in a different combination.

In this connection consider what appellants say in their specification concerning the purpose and function of aluminum in their alloy. They state:

"The aluminum content of the alloys of the present invention is of considerable importance. If the aluminum content exceeds 0.74%, the ductility of the alloy falls and its welding properties are adversely affected due to skin formation. On the other hand, if the aluminum content is substantially less than 0.3% the resistance of the alloy to deformation under stress falls off to a low value (e. g. deformation of more than 1.0% strain after 100 hours at a temperature of 775° C. under a stress of 7.8 tons per square inch)."

This teaching is nowhere "described" by Gresham et al.

There are two places in the Gresham reference in which the examiner and the board purport to find a description of the alloy-composition recited in and forming part of the appealed claims —

First: alloys (f) and (g) of Gresham. The board concluded that alloys (f) and (g) "fall within the claimed ranges" and therefore that "the claimed alloys are old." Despite these statements, elsewhere in the board's opinion we find an admission that "a slight difference in the aluminum content" exists.

The examiner, on the other hand, seems to have had no doubt, when the final rejection was entered, that "technically speaking applicants' claimed alloys embrace novelty over alloys f and g of patentees." As I have indicated previously, it was not until his "Answer" that we find the suggestion for the first time that, while alloys (f) and (g) were "slightly outside" the claims, they are "meant to be representative" of what the claims embrace.

The "twelve nickel alloys" of col. 2, lines 36-69 of Gresham et al. are the second place where the examiner and the board found appellants' invention to be described. The examiner based his final rejection primarily upon those of the "twelve nickel alloys" which the chart attached to the reference shows to have "balance factors" below 16. However, neither the examiner nor the board have pointed to any description of a specific alloy which falls within the constituent limits of the claims and has a balance factor less than 16.

The "balance factor" referred to by appellants and by Gresham et al. is essentially a mathematical statement in which each of the constituents of an alloy is "weighted" as it were to indicate something of its activity in the formation of the final alloy. Both Gresham et al. and applicants assign a "value" of 2 to aluminum in arriving at the "balance factor." Thus, the "weighted" effect of aluminum in producing appellants' alloy is stated mathematically by multiplying the ranges here claimed by 2. Thus the range of 0.3% to 0.74% have values of 0.6 to 1.48 in the "balance factor." Subtracting this range from the claimed "balance factor" of "below approximately 16," (taken as 16) leaves values of 15.4 to 14.52 as the available range for the other constituents of the alloys. Gresham et al. examples (a) and (b) specify a range of aluminum of from 0.48% to 3.75%. When "weighted" by 2 to arrive at the "balance factor," these percentages have values of 0.96 to 7.50. Subtracting these values from the "balance factor" of 16, we find that Gresham et al.'s alloys permit values of 15.04 to 8.50 as the available range for the other constituents of the alloy. It seems apparent, therefore, that except for a very narrow range of theoretical overlap, alloys within the ranges of the rejected claims are necessarily very different from the alloys of Gresham et al.

This analysis places the Gresham et al. teaching in its most favorable light. Actually, in accordance with the data tabulated in the Gresham et al. patent, it was established that the balance factor should be in the range of from 16 to 20 to impart a relatively high degree of creep strength to the alloy in question. The Gresham teaching indicates that an alloy having a balance factor less than approximately 16 is not suitable for use in applications requiring high creep strength.

Appellants point out in their specification:

"Although the development of the alloys described above [Gresham et al.] satisfied certain requirements in the manufacture of equipment ordinarily subject to relatively high temperatures, other difficulties in the field of high temperature design remain unsolved. Thus, for example, the production of the inner walls of jet pipes of modern gas turbine jet engines was hindered by the lack of a suitable material of construction. Such inner walls must be fabricated in a manner which will enable them to withstand repeated, rapid, thermal cycling up to temperatures of the order of 800° C."

"It has been discovered that chromium-nickel base alloys having a particular composition herein described will possess a relatively high degree of ductility in addition to being relatively resistant to the type of thermal cycling mentioned above."

Thus, it seems to me that an objective appraisal of appellants' invention at the time it was made compels one to recognize the problem which existed after Gresham et al. had produced their alloys, and that the alloy entities sought by appellants and by Gresham et al. have entirely different properties resulting from the different properties resulting from the different ranges of the constituents used. Certainly, I fail to see wherein appellants' inventive concept when considered as a whole is in any wise "described" by Gresham et al.

There remains the further consideration of whether the invention of the appealed claims is "obvious" in view of Gresham et al. under the conditions specified in section 103. Here, I think the examiner, the board and the majority have failed to give adequate consideration to the Meetham affidavits filed by appellants. In this connection I agree with appellants that:

"There is nothing in the Board's opinion to indicate that any weight whatever was given to the Meetham affidavits or to the statements of criticality with respect to the 0.74% upper limit for aluminum which are set forth in the specification. The Board states its view to be (R45-6) —

"* * * that the testing of merely two examples and comparparing them to alloy `X' is insufficient to establish that any alloy falling within the broad ranges encompassed by the claims would produce similarly improved results.

"This looks to much less than the whole record before the Board. Apparently the Board thought that the number of examples which the Application discloses is all that need be considered on an issue of critical- ity.* Specifically, the Board's reasoning dismisses as irrelevant both the specification's statement of the reasons why the 0.74% upper limit is critical, and the Meetham Affidavits' test results which taken with the specification's data strongly support what the application says. * *

My consideration of Meetham's affidavits leads me to the conclusion that the results shown by the tests therein set forth were indeed "unobvious." Except as one imparts into the Gresham et al. teachings the narrow limits of aluminum disclosed by appellants and claimed in the appealed claims, one simply does not produce from the Gresham et al. teachings an alloy comparable in properties to the alloys produced by appellants.

The distinct properties which appellants are striving to produce are best stated in terms of an "article" or a "weld" made from the alloy having the constituents as claimed in the appealed claims. When we thus consider these claims we start with the fundamental proposition that except as both parties are concerned with jet engine components, they are striving to produce parts having very different uses and requiring very different physical properties. As appellants point out in their brief:

"Thus, the `person having ordinary skill in the art' of making jet pipes, with the Gresham Reference before him would, in order to obtain Appellants' thermal shock resistance in a welded jet pipe, be obliged to supply the following items of supplementary information from his own background knowledge —

"1. Something unusual would be needed to draw his attention in the first place to a disclosure dealing with quite a different problem and an altogether different set of criteria.

"2. He would need to know that a blade alloy would, when modified in several respects, be useful as a welding alloy for joining the sheets of a jet pipe.

"3. He would have to make the tests necessary to exclude those of the Gresham alloys that contain chromium below 19% and above 23% and cobalt below 12% and above 25%.

"4. He would have to make the tests necessary to exclude those of the Gresham alloys that contain molybdenum below 3.0% and above 8.6%, aluminum below 0.3% and above 0.74%, and titanium below 1.7% and above 2.45%.

"Having come thus far, he would still not have reached the indicated goal as `Alloy X' of the Application shows. Additionally —

"5. He would have to discard the teaching of Gresham that desirable properties disappear when the `balance factor' is below 16, and make the tests necessary to exclude those of the Gresham alloys, otherwise narrowed down as indicated above, that possess a `balance factor' above 16."

By hindsight, it is very easy to assume, as does the majority, that one of ordinary skill in the art would know, understand and utilize appellants' critical percentages of aluminum in the Gresham et al. alloys. However, the articles produced from such an alloy would have properties which would defeat the very purpose for which the Gresham et al. alloys were developed. In my opinion, the differences between the invention of Gresham et al. and the invention of the appealed claims are such that appellants' invention as a whole would not have been obvious at the time it was made to a person having the ordinary skill of the art.

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