delivered the opinion of the court:
The issue in this appeal is whether the invention as embodied in the chemical compounds, defined in appealed claims 6 and 13, would be obvious in view of the teachings of the prior art references of record.1 The claims read as follows:
6. l-Ethyl-3-pyrrolidyl benzilate hydroehoride.
13. l-Ethyl-3-pyrrolidyl benzilate methobromide.
The compounds are disclosed as being active in inhibiting intestinal spasms. The application 2 states that an object of the invention is “to provide novel acetylcholine antagonists having a high degree of activity and satisfactory activity when compared with known compounds currently in use as inhibitors of gastrointestinal motility.”
Claims 6 and 13 stand rejected as “unpatentable over” the teachings of the following references:
Martin et al_ 2,987,517 June 6,1961
Biel [I]_ 2,918,406 Dec. 22,1959
Biel [III]_ 2,918,408 Dec. 22,1959
Ekenstam et al_ 2,792,399 May 14,1957
Bloxam (British)- 483,258 Apr. 14,1938
Ricbter, Organic Chemistry, Vol. 3 at p. 3 (1923).
The two rejections made by the examiner were treated by the Board of Appeals as follows:
Claims * * * 6 * * * [and] 13 * * * have been rejected as unpatentable over each of Biel I (2,918,406) or Biel III (2,918,408) in view of Ricbter and the British patent, and all the claims have further been rejected on these same references with the addition of the Martin et al. patent. The Examiner has also referred to the Ekenstam et al. patent as being cumulative in showing the equivalence of pyrrolidine and piperidine compounds. Inasmuch as the two rejections are quite similar, based upon the same references except for Martin et al., we will consider them together.
According to the examiner,
* * * The primary references [Biel I and Biel III] disclose the corresponding homologous piperidine benzilates while the ancillary art shows the equivalence of pyrrolidine and piperidine compounds (British patent 483,258) and the homology of pyrrolidine and piperidine (Richter’s Organic Chemistry). No invention can be seen in obvious equivalents of the prior art.
In view of the manner in which we dispose of the issue presented here, we also shall consider the rejections together and shall adopt the board’s nomenclature of “Biel I” and “Biel III” in referring to these references.
*1013Because of the nature of the issues presented in appellant’s arguments, a detailed discussion of the chemistry involved in the subject matter sought to be patented and the prior art is not necessary. It ■will suffice to point out that concerning the “primary references,” Biel I discloses antispasmodic specifics (N-ethyl-3-piperidyl benzilate compounds) which have selective action in the stomach; and Biel III discloses antispasmodic specifics (bT-methyl-S-piperidyl benzilate compounds) which have selective action in the colon. The pertinent teachings of the “ancillary references” are as follows: the Richter text teaches that pyrrolidine and piperidine are adjacent members of a homologous series; the British and Martin patents were relied on to show that esters exhibiting antispasmodic activity may contain either the pyrrolidyl or the piperidyl ring; and Ekenstam was regarded “as cumulative in also showing pharmaceutical compounds containing either the pyrrolidyl or the piperidyl ring.”
The structural difference between appellant’s claimed compounds and those disclosed in the primary references is that the nitrogen heterocyclic rings of the claimed compounds are the five-membered pyrrolidyl rings whereas in the Biel compounds the nitrogen hetero-cyclic rings are the six-membered piperidyl rings, as shown in the following formulae:
Considering the closest compounds, m the appealed claims and Biel I, R is ethyl; and in Biel III, R is methyl.
Reference Teachings
Appellant argues that the rejections are improper because of certain teachings found in the primary references.
Biel I states:
Surprisingly, N-ethyl-3-piperidyl benzilate onium salts have a pronounced and selective action in the stomach, and have been found to be especially effective in the treatment of peptic ulcer, without significantly altering normal tonus or motility of the stomach or intestines.
*******
Such unexpected specific activity in the stomach is even more surprising because the next adjacent homolog N-methyl-3-piperidyl benzilate, as the methyl bromide, does not have this specific activity but, instead, is specific for treating spasm of the colon.
*1014Biel III states:
Surprisingly, N-metliyl-3-piperidyl’ benzilate methonium compounds have a unique, highly selective antispasmodic action in the lower part of the gastrointestinal tract, i.e., the colon. * * *
Such unexpected activity in the colon is even more surprising because the next adjacent homolog N-ethyl-3-piperidyl benzilate, as the methyl bromide, does not display activity in the colon but is a selective agent for peptic ulcer in the stomach and duodenum because of its action in suppressing gastric secretion and motility in this area.
For ease of consideration, the teachings of Biel I and III may be set forth as follows:
Mthyl-plperidyl Form of Compounds
1. “have a pronounced and selective action in the stomach” (Beil I)
2. “[are] especially effective in the treatment of peptic ulcer” (Beil I)
3. “[do not] significantly * * * [alter] normal tonus or motility of the stomach or intestines” (Beil I)
4. “[do] not display activity in the colon” (Beil III)
5. “is a selective agent for peptic ulcer in the stomach and duodenum because of its action in suppressing gastric secretion and motility in this area.” (Beil III)
Methyl-piperidyl Form of Compounds
1. “have a unique, highly selective antispasmodic action in the lower part of the gastrointestinal tract, i.e., the colon.” (Beil III)
2. “[are] specific for treating spasms of the colon” (Beil I)
3. “[do] not have * * * [the] specific activity of [Beil I teachings 1, 2 and 3 under Ethyl piperidyl]”
Considering the teachings of the Biel references as a whole, we find that the ethyl-piperidyl form of compound is especially effective in the treatment of peptic ulcer, has selective action as an antispasmodic in the stomach and duodenum, does not display activity in the colon, and does not significantly alter normal tonus or motility of the stomach or intestines.3 The methylpiperidyl form of compound is espe*1015cially effective in tlie treatment of spasm of the colon and does not. display activity in the stomach or duodenum. The reference teachings-convey a direct impression that each of the above forms of compounds operate in mutually exclusive areas of the gastro-intestinal tract as-antispasmodics and that this is due to the influence of ethyl and. methyl groups.
Appellant argues that given the teachings that the ethyl-piperidyl ring form of compound is an antispasmodic only in the stomach and. duodenum and that the methyl-piperidyl ring form of compound is. an antispasmodic only in the lower intestinal tract, it would not be-obvious under the conditions of 35 TJSC 103 that the ethyl-pyrrolidyl ring form of compound would be an antispasmodic in the intestines,, as disclosed in his application.
The rejection here is under 35 USC 103. The rejection is that the differences between the subject matter as a whole and the specific-teachings of Biel I and III in view of the general teachings of the-ancillary references are such that the subject matter would be obvious, to one of ordinary skill in the art. We think as a matter of law that. 35 USC 103 requires that the specific teachings, as well as the general teachings, be evaluated as a whole for what they fairly teach.
In the present case the claimed invention, considering the properties along with the structure, is clearly not obvious in view of the-general teachings. It cannot be argued other than that these teachings are concerned only in a peripheral manner with the claimed invention. The solicitor does not argue otherwise. The specific teachings, on the other hand, are closely concerned with the claimed invention. Absent any reason to the contrary, it is these teachings which a person or ordinary skill in the art would consider most pertinent and accord the most weight. The specific teachings, uncontradicted by the art. of record, disclose that the ethyl-piperidyl form of compounds performs as an antispasmodic in the stomach and duodenum and will not perform as a lower intestinal antispasmodic. As a matter of law and logic we think that it would not be obvious that the ethyl-pyrrolidyl' form of compound would perform as an intestinal antispasmodic. Thus the examiner erred in ignoring the specific teachings of the-primary references, without citing other references containing specific-teachings demonstrating that the specific teachings in Biel I and III could be ignored.
The Affidavits
The board in its decision relied on certain evidence submitted by-appellant as further establishing the obviousness of the claimed compounds. This evidence consisted of affidavits which the board ac*1016cepted as establishing that the compound of claim 13 was “4.8 times as potent as the corresponding piperidyl compound of Biel I.” The board found:
* * * At the best, the compound of claim 13 may exhibit the same property to a somewhat enhanced degree. A mere improvement in the same property does not establish unobviousness of the claimed compounds. See In re Lohr et al., 50 CCPA 1274, 794 O.G. 770, 317 Fed. (2d) 388, 137 USPQ 548, and In re Riden et al., 50 CCPA 1411, 796 O.G. 863, 318 Fed. (2d) 761, 138 USPQ 112.
■ In its decision on appellant’s petition for rehearing the board elaborated further:
* * * We can perceive nothing unexpected in finding that the claimed compounds have colonic antispasmodie activity, like the compounds of Biel III.
Equally important is the fact that, despite the disclaimer by Biel, the ethyl substituted piperidyl compound of the Biel I patent has been shown by appellant to actually possess colonic antispasmodie activity. In short, the ethyl substituted compound of the reference actually exhibits the same type of activity as the compounds claimed by appellant in claims 6 and 13.
Appellant reasons the “discovery” (that the ethyl-piperidyl ring form of compound of Biel I does in fact possess intestinal antispasmodie properties) is his and the board in effect used his own discovery against him.
The solicitor in his brief does not dispute that the discovery set forth in the affidavits was used against appellant. Instead it is argued:
* * * While appellant claims credit for this discovery, it might be noted that the examiner suspected such activity and required comparison tests with the reference compounds (Br-6). Therefore, the true situation came to light, and as might be suspected, substitution of closely related groups results in a difference in potency rather than a change in activity. Thus, appellant’s compounds do not have an activity which would be unexpected from the reference compounds. [Emphasis added.]
Prior to considering whether the board erred in its use of the affidavit evidence, it must be decided whether the solicitor’s contentions are correct. We have examined page 6 of the appellant’s brief cited by the solicitor and do not find support for the solicitor’s statement that the discovery was a direct result of the examiner’s suspicions. At page 6 of his brief appellant argues:
Thirdly, because the Examiner would not pay any attention to the teaching of the reference that the exactly corresponding piperidyl compounds of the prior art do not have intestinal antispasmodie activity, the appellant made and tested the closest representative one of them and presented actual comparative data as to the compound of Claim 13 against the corresponding prior art compound having a piperidyl ring instead of the pyrrolidyl ring. In this test the appellant’s compound was 480% as potent in decreasing intestinal spasm * * *.
Appellant argues throughout his brief that the examiner ignored the specific teachings of Biel I and III. Additionally, he argues that *1017'-fclie affidavits were submitted, in effect, as a last resort and that no affidavits were needed to overcome the rejection in view of the teachings in Biel I and III.
. We think the record supports the finding that the examiner ignored the statements in Biel I and Biel III, supra, which alleged the methyl ■and not the ethyl type of compounds were lower intestinal antispasmodics, and that the ethyl was effective in the stomach and duodenum. In his answer the examiner stated:
The argument that the “subject matter as a whole” under 35 U.S.O. 103 Includes the compound and its utility is considered to be without merit. (In re Schmidt 130 U.S.P.Q. 404; in re Larsen 130 U.S.P.Q. 209 at page 212, 2nd column). A bare compound will anticipate a claim to same regardless of the utility disclosed for it.
The claimed compounds haying been shown to be expected functional equivalents of the prior art, they are deemed obvious under 35 U.S.O. 103.
Ignoring the statement as to “anticipation,” 4 it is reasonably clear that the examiner considered only the difference in strueiyures between the claimed compounds and the prior art compounds.
Appellant was entitled to have differences between the claimed invention, the subject matter as a whole, and the prior art references of record evaluated. In re Papesch, 50 CCPA 1084, 315 F. 2d 381, 137 USPQ 43. This the examiner failed to do. We note that the examiner’s answer is dated December 13, 1961 and our decision in Papesch is dated March 20, 1963.
A person of ordinary skill in the art would consider all of the teachings in Biel I and III as well as the general teachings in the other references. It is not believed that he would consider only the structures of the compounds or would disbelieve specific teachings and rely only on less helpful general statements.
The record thus does not support the solicitor’s position that the examiner in fact “suspected” that the compounds in Biel I possessed intestinal antispasmodic potency. Moreover, as a matter of law under 35 USC 103, the examiner must substantiate his “suspicions” 5 on the basis of facts drawn from proper prior art. The issue to be resolved requires more than “suspicions;” it requires facts. As the Supreme Court recently stated concerning section 103:
* * * the § 103 * * * condition [nonobviousness], when followed realistically, will permit a more practical test of patentability. The emphasis on nonob-*1018viousness is one of inquiry, not quality and, as such, comports with the constitutional strictures.
* * * the § 103 condition * * * 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 he ascertained; and the level of ordinary skill in the pertinent art resolved. Against this: background, the obviousness of the subject matter is determined. [Emphasis-added.] Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459, 466-67.6
The provisions of section 103 must be followed realistically to develop the factual bachgrou/ncl against which the section 103 determination must be made. All of the facts mnst be considered and it is-not realistic:
* * * within the framework of section 103 to pick and choose from any one reference only so much of it as will support a given position, to the exclusion, of other parts necessary to the full appreciation of what such reference fairly suggests to one of ordinary skill in the art. In re Wesslau, 53 CCPA 746, 353 F. 2d 238, 147 USPQ 391 at 393.
We now return to the question of whether the board erred in its use of the affidavit evidence. The record shows that appellant was unable to convince the examiner that the specific teachings in Biel I and III must be considered. He then conducted experiments testing the potency of the compound of claim 13 and an ethyl-piperidyl ring form of compound from Biel I as an intestinal antispasmodic. The resulting data, demonstrating that the Biel I compound did in fact possess potency but that the compound of claim 13 was 4.8 times as potent as Biel I, was submitted in an affidavit. The examiner and the board reasoned that this evidence established that appellant’s compounds-were merely more effective than the Biel I compound as a colonic intestinal antispasmodic.7 This reasoning ignores the critical fact that the prior art teaches that the compound in Biel I was antispasmodic for the stomach and duodenum.
*1019It is clear tRat the affidavit evidence is not prior art under 35 USC 103 and may not be used alone or in combination with prior art references to support a rejection under that section. In re Lunsford,, 53 CCPA 986, 357 F. 2d 380, 148 USPQ 716. We think the board erred in its use of the affidavit evidence.
Moreover, the record shows that this is not the type of case under 35 USC 103 where affidavits are needed.8 The prior art taught that the ethyl-piperidyl form of compound was effective as an antispasmodic for the stomach and duodenum. Appellant therefore did not have to submit evidence that the potency of the claimed compounds was unexpectedly superior to the potency of the prior art compounds in the intestines. He was entitled to rely on the fact that the most specific teachings cited by the examiner, uncontradicted by the art of record, led away from what appellant claimed.
The decision of the board is reversed.
WoRlet, C. J., did not participate.
An appeal as to claims Í — 5, 12 and-19 hás beeii withdrawn;
Filed May 16, 1960, Serial No. 29,156, entitled “l-Substituted-3-PyrroIidyl Benzilates.V
The Biel compounds are effective only as to certain spasms and only in certain areas of the gastro-intestinal tract, meaning the stomach and intestines. This pinpointing of areas of effectiveness and the teachings as to the type of spasms controlled present some difficulties in interpretation of their teachings. Biel I states the compounds there have “selective action in the stomach” and are “especially effective in the treatment of peptic ulcer,” which may occur in the esophagus, stomach or duodenum. They do not display activity in the colon, the large intestines. While Biel I compounds are effective in the duodenum, the initial portion of the small intestines, for treatment of peptic ulcer, they do not significantly alter “normal tonus or motility of the stomach or intestines However, they are effective in reducing “either involuntary muscle or nervous spasm.” For the purposes of this ex parte appeal, we think Biel I compounds are effective anti-spasmodies for the treatment of spasms associated with peptic ulcer in the stomach and duodenum; we also think that these compounds do not display such activity in the remaining portion of the intestines, because of the emphasis on peptic ulcer and the fact that the duodenum is but a small portion of the total intestines (approximately 1/25, the duodenum being 8-10 inches, the small intestines 20 feet, and the large intestines 5 feet).
There is no explanation for this statement in the record. It is clear from the record that there was no “anticipation” rejection, i.e., fully met in a single reference, (35 USC 102 (a)) before the board nor was such a rejection presented or argued in this court. The claims stand rejected on the broader ground that the subject matter defined in the appealed claims is obvious under the terms of 35 USC 103.
We distinguish between “suspicions,” “hunches,” etc., and statements of fact concerning the state of the prior art, advanced by the parties during prosecution about which there is no disagreement or challenge.
The Supreme Court continues in its opinion, recognizing other considerations relevant to the inquiry under section 103 :
Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented. As indicia of obviousness or nonobviousness, these inquiries may have relevancy. See Note, Subtests of “Nonobviousness,” 112 U. Pa. L. Rev. 1169 (1964).
While the board refers to colonic antispasmodic activity, the affidavits cited by the board show tests made concerning the ileum, which is 3/5 of the small intestine.
The hoard also expressed dissatisfaction with the data presented In the affidavits. As we do not rely on that evidence, we need not consider the board’s comments.