RICH, Judge.
This appeal is from the decision of the United States Patent and Trademark Office (PTO) Board of Appeals (board) sustaining the rejection of claims 1-3, 6, and 7 of application serial No. 821,360, filed August 3, 1977, for “Elastomeric Sealant Composition,” on the ground of double patenting and also on a theory of abandonment by assignment, citing 35 U.S.C. § 102(c). We affirm the double patenting rejection and do not reach the abandonment issue.
Background
The appellant-inventors, Van Ornum and Stang, in addition to filing the application [938]*938at bar, had already had issued to their assignees two United States Patents on puncture sealant compositions for vehicle tires: No. 3,935,893, issued Feb. 3, 1976 (the ’893 patent), and No. 4,113,799, issued Sept. 12, 1978 (the ’799 patent). It has been of significance in this case that the ’893 patent issued to General Motors Corporation as the result of an assignment, recorded in the PTO on the filing date of the application for that patent. The present application and the ’799 patent have been assigned to Rocket Research Corporation, by change of name now Rockcor, Inc., the real party in interest in this appeal, the assignment also being recorded in the PTO.
The application on appeal was filed under 35 U.S.C. § 121 as a voluntary division of the application that matured into the ’799 patent, and all claims stand rejected on the ground of double patenting, said to be of the “obviousness type,” the rejection being predicated on the claims of both the ’893 and ’799 patents.
The board opinion explains its views on the relationship of the claims of the two patents to the claims on appeal as follows:
Before we discuss the rejections, it would be well to set forth an analysis of the various inventions being claimed in the two patents and the present application. As can be seen, all three relate to sealing compositions, which are to be used in the same manner. The claims of Patent No. 3,935,893 set forth the composition in its most detailed form, describing six ingredients, consisting of the high molecular weight and low molecular weight butyl rubbers, liquid polybutylene, partially hydrogenated elastomeric block co-polymer, carbon black and cross-linking agents. The ratio of high molecular weight butyl rubber to low molecular weight butyl rubber is specifically described as being 60 to 40. Patent No. 4,113,799 is broader in that it sets forth only the high molecular weight and low molecular weight butyl rubber as well as a “tackifier.” As can be seen from the patent specification, liquid polybutylene is a preferred tackifier. The claims define a broad range of 35-65 for the high molecular weight rubber to 65-35 low molecular weight butyl rubber. The claims in the present application describe the same high molecular and low molecular weight butyl rubber composition which is used with the tackifier. However, the ratio of high molecular weight to low molecular weight butyl rubber is set forth in a broader range of 20-60 high molecular weight rubber to 80-40 low molecular weight butyl rubber. Also, the specification teaches that liquid polybutylene is the preferred tackifier. [Our emphasis.]
Appellants’ brief says that the claims on appeal are identical to claims which were in the application for the ’799 patent which were rejected therein on the ground of double patenting in view of the ’893 patent claims and that, because the Manual of Patent Examining Procedure (MPEP) § 1490 states that a terminal disclaimer cannot be directed to particular claims but only to an entire patent, they were divided out and placed in the divisional application on appeal. The ’799 patent then went to issue with claims which had been allowed. The prosecution of the broader claims now before us was continued in a divisional application. With further reference to the relation of the claims on appeal to the claims of their patents, appellants’ view is as follows:
The claims of the present applications [sic] are, by definition, generic to the species claims of the two patents * * * relied upon by the Examiner in support of the double patenting rejection * * *. The broad generic claims of the application comprehend both the constituents and specific range limitations of the claims of both patents. The Board of Appeals implicitly acknowledges the genus species relation by treating the appealed claims as an attempt to claim" broadly that which had been previously described in more detail in the claims of the two patents.
In prosecuting these “generic” claims in the application before us, appellants sought [939]*939to overcome the double patenting rejection by filing a terminal disclaimer under 35 U.S.C. § 253, second paragraph.1 The first disclaimer filed was criticized as not in proper form and a second one was filed. The PTO rule or regulation on terminal disclaimers, which gives rise to the issue before us, is 37 CPR 1.321 which reads:
§ 1.321 Statutory disclaimer.
(a) A disclaimer under 35 U.S.C. 253 must identify the patent and the claim or claims which are disclaimed, and be signed by the person making the disclaimer, who shall state therein the extent of his interest in the patent. A disclaimer which is not a disclaimer of a complete claim or claims may be refused recordation. A notice of the disclaimer is published in the Official Gazette and attached to the printed copies of the specification. In like manner any patentee or applicant may disclaim or dedicate to the public the entire term, or any terminal part of the term, of the patent granted or to be granted.
(b) A terminal disclaimer, when filed in an application to obviate a double patenting rejection, must include a provision that any patent granted on that application shall be enforceable only for and during such period that said patent is commonly owned with the application or patent which formed the basis for the rejection. [Our emphasis.]
The legal problem which has been paramount in this case arose because appellants would not, because they could not, file a disclaimer complying with paragraph (b) of the rule because their ’893 patent, on the claims of which the double patenting rejection was in part based, had been assigned to General Motors. The PTO therefore ruled that the disclaimer was unacceptable and that the double patenting rejection would have to stand. Appellants therefore attack the rule.
With respect to the double patenting rejection, appellants make two principal contentions: (1) the appealed claims, considering their relation to the claims of the two patents, were not properly rejected for double patenting; (2) 37 CFR 1.321(b) is invalid, (a) as beyond the rulemaking authority of the Commissioner of Patents and Trademarks under 35 U.S.C. § 6(a) and (b) because it is contrary to statutory and case law.
OPINION
The Facts Adequately Support a Rejection Based on “Double Patenting”
We necessarily begin by considering whether there was a proper “double patenting” rejection. We approach that question by stating our view of the essential facts as gleaned from our own study of the record.
Appellants Van Ornum and Stang are the identical inventorship entity involved in all of the patents and application here involved, notwithstanding the reversal of the order of their names on the two issued patents so that they have been referred to as the Stang et al. and Van Ornum et al. patents. They filed their first application on July 15,1974, and simultaneously recorded their assignment thereof to General Motors. The ’893 patent issued thereon is entitled “Self-Sealing Vehicle Tire and Sealant Composition.” The object of the invention is to render tubeless automobile tires self sealing after being punctured, desirably maintaining that capacity over a wide possible service temperature range of -20 °F. to 270 °F. The inventors were not pioneers in the field and begin their disclosure by referring to three prior sealant patents. They say they began their work by “screening a number of commercial sealants” containing curable butyl rubber which lacked sufficient tack and strength at high temperatures. The first example of the patent, which contains two specific examples, discloses that appellants began with a “corn[940]*940mercially available sealant composition,” the makeup of which is tabulated, to which they added a liquid tackifier and a “block copolymer.” Going into more detail, the tabulation of the old sealant composition shows that it contained a by-weight mixture of 60 parts of high molecular weight butyl rubber and 40 parts of low molecular weight butyl rubber, a quantity of mixed carbon blacks of three different grades, a cross-linking agent, and a large amount of toluene as solvent. To this was added, as tackifier, a liquid copolymer consisting of 98% butylene and 2% isobutylene which was “a commercial product available under the trade name Tndopol H-300’.” Also added was a block copolymer “of the A-B-A type wherein the A blocks were formed of polystyrene, [and] the B blocks were polymeric segments of isoprene and some higher carbon chain length conjugated dienes,” partially hydrogenated. This material “was obtained under the trade designation ‘Kraton G-6500’,” said to be about 68% by weight polyisoprene.
This condensed description should suffice to give meaning to the four claims of ’893, which are of quite limited scope, claim 1 being exemplary:
1. A sealant composition for use in a vehicle tire to seal punctures therein up to about one-quarter inch in diameter formed in the operation of said tire, said composition consisting essentially of, by weight,
10 to 15 parts of a butyl rubber having an average molecular weight in the range of about 100,000 to 300,000,
6 to 10 parts of a butyl rubber having an average molecular weight in the range of about 10,000 to 30,000,
60 to 65 parts of a liquid polybutylene having an average molecular weight of about 500 to 5,000,
5 parts of a partially hydrogenated thermoplastic elastomeric block copolymer having the general molecular configuration A-(B-A)l-5 wherein, prior to hydrogenation, each A is a monovinyl arene polymer block and each B is a conjugated diene polymer block, and only said diene block(s) are hydrogenated.
5 to 17 parts carbon black, and small but suitable amounts of cross-linking agents suitable for curing butyl rubbers.
All four claims are of the “consisting essentially of” type. Claims 1 and 2 are directed to “A sealant” and claims 3 and 4 to “A self-sealing rubber tire” containing sealant as defined in claims 1 and 2, respectively. Claims 2 and 4 differ from claims 1 and 3 in that they specify the monovinyl arene to be styrene and limit the “diene polymer” as being “from the group consisting of butadiene and isoprene.” So much for the nature of patent ’893. We turn now to the other patent. .
One year after filing the application for their ’893 patent and assigning it to General Motors, appellants filed, on July 14, 1975, while their first application was still pending, their application which matured into the ’799 patent. The two applications are unrelated in that the second patent makes no reference to the first application. It also has a different assignee, Rockcor, Inc. We do not have the file history of this patent in the record so all we know about it is what is in the patent and what we have been told about it, which is minimal. Essentially all of the disclosure of the ’893 patent is contained in it but in rewritten and considerably amplified form. Instead of two specific examples there are eight, so designated, plus a table of eight somewhat related specific compositions, and two added drawings one of which is a bicycle tire and the other a graph. One item of added disclosure relates to the ratio of high to low molecular weight butyl rubbers which form an important ingredient of all the sealants. The general disclosure is that the weight ratio “may vary from 20/80 to 60/40.” A more specific disclosure is that “a sealant composition having a ratio of high molecular weight to low molecular weight butyl rubber between about 35/65 to 45/55 has unexpectedly superior properties.” The patent specification goes on to explain that superiority in detail, which we need not consider, [941]*941and was evidently enough to persuade the examiner that this narrower ratio range was a patentable discovery since it became the subject matter of the claims which appear in the patent. Claim 1, the only independent claim, exemplifies this:
1. A sealant composition comprising a reinforced partially cross-linked matrix comprising a high average molecular weight butyl rubber having a molecular weight in the range of approximately 100,000 to 400,000 and a low average molecular weight butyl rubber having a molecular weight in the range of approximately 10,000 to 40,000, in a ratio of high to low molecular weight butyl rubber of between about 35/65 and 45/55, in admixture with a tackifier present in an amount between about 55 and 70 weight % of the composition.
The remaining claims of the ’799 patent are the dependent claims 2-7 adding further limitations. For example, illustrative of the kind of added disclosure in this patent over that in patent ’893, claim 6 states that the “tackifier” of claim 1 may be selected from a group consisting of nine broadly-named subgroups of compounds which will be found listed in the specification, e.g., one such subgroup is “alkyl aromatics.” Compared with the disclosure of patent ’893, practically every component of the sealant there described is the subject of an augmented disclosure of possible variants in patent ’799. Another interesting feature of the prosecution leading to patent ’799 is a declaration of Van Ornum under 37 CFR 1.131 which the attorney stated (our emphasis) “establishes that the genus claimed in this application on appeal [to the board] was reduced to practice before the filing date of U.S.P. 3,935,893.” One statement in that declaration is that 5 of the 8 examples in the table of compositions in the ’799 patent, repeated in the division therefrom now before us, were physically tested prior to the filing date of the first patent, ’893, July 14, 1974.
We now approach the application at bar. We are told that the claims on appeal were originally presented in the application for patent ’799 but were rejected therein for double patenting in view of the claims which now appear in patent ’893.2 Appellants, thinking of avoiding that rejection with a terminal disclaimer, discovered that they would have to disclaim a terminal portion of the term of any patent they might get with respect to all the claims in their application, including claims 1-7 which were allowable. They avoided this sacrifice by dividing out the claims on appeal into the application before us, without changing the specification, and patent ’799 was issued for its full term. Application serial No. 821,360 is the divisional application they filed on Aug. 3, 1977, here on appeal. Therein they received a final rejection for double patenting in view of the claims of patent ’893, and also in view of the allowed claims of the parent application which matured into patent ’799. The examiner issued this rejection after receiving a “Disclosure Statement” from appellants’ attorney containing the following (the omissions being merely page references):
U. S. Patent 3,935,893 discloses and claims a tire sealant composition comprising a high molecular weight butyl rubber, a low molecular weight butyl rubber and a partially hydrogenated block copolymer. The block copolymer is stated to be absolutely essential * * * and must be present in at least 4% by weight * * *. In contrast, the invention disclosed and claimed herein does not require a partially hydrogenated block copolymer, since applicants have discovered that certain compositions make good sealants for a variety of uses including bicycle tire sealants whether or not they include such [942]*942block copolymers. * * * This difference requires the conclusion that the invention claimed in U. S. Patent 3,935,893 and the invention claimed in this application are patentably distinct.
The examiner disagreed with appellants’ conclusion and so do we, at least for present purposes as more fully explained later. For the present, we note that this “absolutely essential” argument remains one of appellants’ principal points and requires closer examination. As earlier stated, the ’893 patent’s stated objective is a sealant effective in a vehicle tire over a wide temperature range of from -20° to 270°. The statement in the specification reads as follows:
* * * we have found that it is absolutely essential that it [block copolymer] be employed in the subject sealant composition if the sealant is to have suitable strength, particularly over the wide temperature range involved, to effectively retain the air under pressure in a punctured tire.
In the amplified disclosure of patent ’799, appellants take a different approach to the whole situation. The whole pitch appears to be to justify broader — even “generic”— claims. The sealant is now not just for automobile tires; it may be applicable to less severe uses, says the disclosure, such as bicycle tires, tire patches, auto sealant, roofing sealant, caulking compound, general household sealant “and others.” Partially hydrogenated block copolymer is no longer “absolutely essential”; “To aid in maintaining sufficient tackiness and thermal stability at elevated temperatures, * * * [it] may be included up to about 10 wt. % of the composition * * *.” (Emphasis ours.) It is in fact included in every one of the eight formulas of the table (4.75-5.01 wt. %) but omitted from all of the patent claims except 4 and 5 and, of course, from all of the claims on appeal. In sum, considering the position taken in appellants’ second patent, the “iffy” nature of the statement in their first patent, and the evident fact that nothing in particular is suggested to be done to make it possible to omit the block copolymer component, appellants’ position that putting it in or leaving it out suffices to create a “patentable distinction” between claims is untenable. This difference is presented in patent ’799 as an option. It appears to us that, in the course of their further investigations of their sealants, appellants simply found that their “absolutely essential” statement in their ’893 patent was untrue or an exaggeration or a misunderstanding.
Our broader problem, however, is to decide whether on all the facts the PTO had good ground for making a double patenting rejection, and, if so, what kind of a double patenting rejection. Following this court’s decision in In re Zickendraht, 50 CCPA 1529, 319 F.2d 225, 138 USPQ 22 (1963), wherein it was suggested to the bar that terminal disclaimers might overcome some double patenting rejections, somewhat more than thirty appeals have been decided on that question, each on its own facts. Out of these decisions there evolved a distinction between “same invention” type cases and “obviousness type” cases. By the time of In re Vogel, 57 CCPA 920, 422 F.2d 438, 164 USPQ 617 (1970), this court had decided that “same invention” meant claims to identical subject matter, i.e., that a claim limitation to halogen was not the same as a claim to chlorine and that meat was not the same as pork, or beef, but that a claim to a yard and a claim to 36 inches or three feet were for the same invention despite their differences in wording. The significance of that was that the court had also decided that 35 U.S.C. § 101 precluded two patents for the same invention, wherefore a terminal disclaimer could be of no help. In re Griswold, 53 CCPA 1565, 365 F.2d 834, 150 USPQ 804 (1966). On the other hand, numerous cases were considered in which application claims were directed to mere obvious modifications of, or improvements on, inventions defined in the claims of patents already issued to the same inventors, or to common assignees, and it had been decided that they might be allowed to go to patent if the applicants filed terminal disclaimers. We classified these as “obviousness type double patenting.” This latter classification has, in the course of time, come, somewhat loosely, to indicate any “double patenting” situation [943]*943other than one of the “same invention” type, although it is not altogether appropriate in cases such as this where, as appellants argue with some justification, their present claims bear a genus-species relation to the claims in the two issued patents, the application claims being generic and the patent claims specific. To illuminate this situation we now quote the only independent claim in the application at bar:
1. A sealant composition comprising a reinforced partially cross-linked matrix comprising a high average molecular weight butyl rubber having a molecular weight in the range of approximately 100,000 to 400,000 and a low average molecular weight butyl rubber having a molecular weight in the range of approximately 10,000-40,000 in a ratio of high to low molecular weight butyl rubber of between about 20/80 and 60/40, in admixture with a tackifier present in an amount between about 55 and 70 weight % of the composition.
The remaining claims on appeal, 2, 3, 6, and 7, are all dependent claims and are identical with correspondingly numbered claims of patent ’799. That being so, comparison of the above application claim 1 with patent ’799 claim 1, previously quoted, shows that the only difference between the claims on appeal and the claims of the ’799 patent resides in the recited ratio of high to low molecular weight butyl rubber, as follows:
application: between 20/80 and 60/40 patent ’799: between 35/65 and 45/55.
The former ratios are broader and inclusive of the latter and are in that sense “generic” and cause the application claims to dominate the ’799 patent claims. They also, of course, dominate the claims of the ’893 patent. Furthermore, the ratios of the claims on appeal are disclosed in the ’799 patent, and these claims were found unpatentable therein, for double patenting, over the claims of the earlier ’893 patent. Consider that claim 1 of patent ’893 recited 10-15 parts of a high weight butyl and 6-10 parts of a low weight butyl rubber, limitations which were exemplified in the express 60/40 ratio of Example 1 and the 14.25 parts high to 9.5 parts low of Example 2, which calculates out to a 60/40 ratio.
This court has encountered this genus-species situation before and considered the obviousness problem. In Vogel, supra, this court said in 1970:
We recognize that it is most difficult, if not meaningless, to try to say what is or is not an obvious variation of a claim. A claim is a group of words defining only the boundary of the patent monopoly. It may not describe any physical thing and indeed may encompass physical things not yet dreamed of. How can it be obvious or not obvious to modify a legal boundary? The disclosure, however, sets forth at least one tangible embodiment within the claim, and it is less difficult and more meaningful to judge whether that thing has been modified in an obvious manner. It must be noted that this use of the disclosure is not in contravention of the cases forbidding its use as prior art, nor is it applying the patent as a reference under 35 USC 103, since only the disclosure of the invention claimed in the patent may be examined.
Again, in In re Braithwaite, 54 CCPA 1604, 379 F.2d 606, 154 USPQ 38 (1967), this court would have held a situation much like the present to require sustaining a double patenting rejection except for the existence of an effective terminal disclaimer. Broader dominating claims were in the continuation-in-part application on appeal and were generic to a claim of an issued patent.
In re Schneller, 55 CCPA 1375, 397 F.2d 350, 158 USPQ 210 (1968), is also pertinent because it involved a voluntary divisional application of which an alleged continuation was before this court with a rejection for double patenting of claims which would have continued patent protection on the preferred embodiment of an invention already patented. What we said in that case applies to the facts here:
The fundamental reason for the rule [against “double patenting”] is to prevent unjustified timewise extension of the right to exclude granted by a patent no [944]*944matter how the extension is brought about. •i* sfc ¿k :$■
* * * if appellant were now to prevail, the end result would be the grant of another patent effectively extending the time during which he may exclude others from practicing an invention which is disclosed and claimed in his issued patent.
This would necessarily be so because of the fact, not only admitted but urged by appellants, that the claims on appeal are generic to — that is to say they dominate — the inventions claimed in both of the patents for which they applied.
On careful review of all the facts of record, we therefore hold that the double patenting rejection of the appealed claims was fully justified. Appellants, in the course of expanding their first application to disclose enough more by way of details, alternatives, and additional uses to support the broad, dominating, “generic” claims here on appeal, have disclosed no additional invention or discovery other than what has already been claimed in patent ’799, as above explained. There is a significant difference between justifying the broadening of claims and disclosing additional inventions.
Since the application claims, compared to the claims of either of the patents cited to support the rejection, do not present a “same invention” type of double patenting situation, the next question is whether appellants can overcome the rejection by filing a terminal disclaimer.
Appellants’ Terminal Disclaimer Is Unacceptable
Appellants concede that the terminal disclaimers they have filed do not comply with 37 CFR 1.321(b), supra, because they are unable to comply with it, having assigned all right, title and interest in the invention on which patent ’893 issued to General Motors Corporation. Their assignment was executed on July 5, 1974, filed with the application, and recorded, in the PTO on July 15, 1974. The parent of the application at bar, serial No. 595,351, filed July 14, 1975, together with all right, title and interest in and to the invention therein described, having been assigned to Rocket Research Corporation, by change of name Rockcor, Inc., that assignee necessarily owns the invention at bar, described and claimed in the divisional application based thereon. Neither appellants nor Rockcor, therefore, can give the undertaking required by § 1.321(b) that any patent granted on the divisional application will be commonly owned with the ’893 patent, “which formed the basis of the [double patenting] rejection,” to quote § 1.321(b).
Since appellants cannot comply with the regulation, they have, perforce, taken the only course left open to them short of surrender and challenge the validity of the regulation.
37 CFR 1.321(b) is a Valid Regulation
Professor Chisum in his recent textbook Patents (1981) includes as Chapter 9, “Double Patenting,” the most complete, critical summary of that law we have seen. On the subject at hand, he says (§ 9.04[2][b]:
[ii] — Harassment by Multiple Assignees. Even though both patents are issued to the same patentee or assignee, it [is] possible that ownership of the two will be divided by later transfers and assignments. The possibility of multiple suits against an infringer by assignees of related patents has long been recognized as one of the concerns behind the doctrine of double patenting.15
He then discusses two cases in this court, in one of which we had played down the importance of harassment by multiple assignees as unlikely, and continues:
The risk of harassment by multiple assignment can be eliminated entirely by the terms of the disclaimer. In a footnote to In re Griswold (1966),20 the court quoted a disclaimer which included a provision that the second patent would “be [945]*945enforceable only for and during such period that the legal title to said patent and to such right to recover shall be the same respectively as” the first patent. The court indicated that this was “ingenious” and “an imaginative solution to one of the more theoretical objections to double patenting, split ownership of two patents and potential harassment.”
After Griswold, the Patent Office amended rule 321 to require terminal disclaimers to contain such a “non-alienation” agreement.21 Apparently no court decision has yet ruled on whether the Patent Office may properly impose such a requirement.
We are now presented with the necessity of making that ruling. In passing, we note that the language of paragraph (b) of the rule is precisely that used in the Griswold terminal disclaimer.
Appellants have elaborated several arguments to support their contention that § 1.321(b) is invalid. We find none of them persuasive and will discuss them seriatim.
The regulation was, of course, promulgated by the Commissioner of Patents and Trademarks and it is contended that it was beyond the authority granted him by the pertinent statute, 35 U.S.C. § 6(a), which gives him the right, “subject to the approval of the Secretary of Commerce, to establish regulations, not inconsistent with law, for the conduct of proceedings in the Patent and Trademark Office.” Appellants say the regulation is “invalid on its face” but they do not explain why beyond contending it is “substantive and not procedural.” We can give no weight to that contention. True, the rule is substantive in that it relates to a condition under which a patent will be granted which otherwise would have to be denied for double patenting. Much of the content of the PTO rules is “substantive” in this respect. The regulation clearly relates to application processing within the PTO in a manner consistent with statutory and case law, which is its principal business. In this connection, we here note briefly the history of the development of the regulation.
We noted above the long line of double patenting decisions in this court which extended over many years. Throughout the period, the theory of harassment by multiple assignees underlay double patenting, at least in part, and now and again came up for discussion. It had long been thought about in the PTO and, no doubt, by some members of the patent bar. In fact, it went back at least to Underwood v. Gerber, 149 U.S. 224, 13 S.Ct. 854, 37 L.Ed. 710 (1893), discussed in Ghisum, supra, § 9.02[4][5]. As this court plowed through its double patenting cases, it came upon the apparently bar-initiated progenitor of § 1.321(b) in Griswold, as noted in the passage quoted from Chisum, supra. That was in 1966. On December 31, 1970, the Commissioner published a Notice of Proposed Rule Making (35 Fed.Reg. 20012), dated December 24, proposing the precursor of § 1.321, in which he said:
The proposed revision of § 1.321 and proposed new § 3.53 brings into the rules a current procedure not based on rule. This provision would prevent harassment of an alleged infringer by multiple parties due to subsequent different ownership of multiple patents granted as the result of filing a terminal disclaimer to overcome a double patenting rejection.
The language of proposed § 3.53 [a new form] is substantially the form which met with the approval of the Court of Customs and Patent Appeals in footnote 5 of In re Griswold, 365 F.2d 834; 150 USPQ 805; 53 CCPA 1565. [Emphasis ours.]
It thus appears that the “current procedure” was adopted sometime after Griswold and had been in effect for some time by the end of 1970. Pursuant to the Notice, a hearing was held on February 19, 1971, comments were received, and the proposed rule was amended in response thereto. The new rules were promulgated and published in 36 Fed.Reg. 7312 with an effective date of April 30, 1971. Among comments re[946]*946ceived in this rulemaking process was one from the American Patent Law Association whose Board of Managers suggested changes and adopted a resolution specifically approving the non-alienation provision. Jan.-Feb. 1971 APLA Bull. 164-165. Thus, the challenged rule has been in effect for more than a decade and this is the first attack on it which has come to our attention.
Appellants say the regulation is contrary to the patent statutes, referencing only the introductory words of § 102, and in derogation of their statutory right to a patent. They also quote a passage from In re Stempel, 44 CCPA 820, 241 F.2d 755, 113 USPQ 77 (1957). There are, and always have been, various legal obstacles that may prevent applicants from obtaining patents which are not spelled out in the statutes. Consider, for example, that for over 160 years all of the law now subsumed under the nonobviousness provision of § 103 was judge-made and without statutory basis. As we have often pointed out, all of obviousness-type double patenting law is still only case law. We think this a sufficient answer to the argument that the regulation is contrary to statute or without statutory foundation.
As for Stempel, of course the case had nothing to do with double patenting, but with what showing must be made under Rule 131 to antedate a reference. It is true that the present writer said in that context that “The patent statutes give to inventors the right to a patent upon compliance with their provisions,” and in that context the statutes were applicable; but in the present context, where the denial of a patent is on a ground with which the statutes have nothing to do, those words are simply too general and cannot be applied. It would be nice if every statement made in an opinion could be automatically tested against every conceivable application to determine whether an exception or two should be noted; but that is a Utopian revery. Precedents are of value for what they decide, not for every sentence they contain. Stempel does not apply here. On another occasion this court has found it necessary to qualify broad language in Stempel found inappropriate to unforeseen circumstances. See In re Tanczyn, 52 CCPA 1630, 1633, 347 F.2d 830, 831, 146 USPQ 298, 301 (1965). It happens all the time; it is in this manner that case law gets refined.
Appellants contend that § 1.321(b) is contrary to or inconsistent with “case law.” The argument is based on two decisions by this court rendered during the formative stages of the law of double patenting and the law of terminal disclaimers as a way to avoid that ground of rejection. They are, in order of their decision, In re Robeson, 51 CCPA 1271, 331 F.2d 610, 141 USPQ 485 (1964), and In re Jentoft, 55 CCPA 1026, 392 F.2d 633, 157 USPQ 363 (1968). The regulation, as we said, was adopted later, in 1971, the PTO and the bar then being quite familiar with those decisions.
In June 1963, this writer, concurring, suggested in Zickendraht, supra, that applicants might make use of terminal disclaimers to avoid obviousness-type double patenting rejections. Robeson came to the court the very next year. Some of the claims had been rejected for double patenting as directed to an obvious variation of what was claimed in a Robeson patent. The application at bar was a continuation-in-part of an application which would have issued with the patent had it not become involved in an interference. Appellant conceded the obviousness issue but argued that he had overcome the rejection by filing a terminal disclaimer. The court began its discussion by commenting that “Whether a terminal disclaimer can overcome the objections to double patenting has been the subject of much discussion.” The opinion then referred to the fact that the rule that there should be only one patent for one invention was as old as Miller v. Eagle, 151 U.S. 186, 14 S.Ct. 310, 38 L.Ed. 121 (1894), and had been followed by this court in In re Siu, 42 CCPA 864, 222 F.2d 267, 105 USPQ 428 (1955), a much debated case in the early days of terminal disclaimer decisions, which really was a “same invention” type of case not curable by terminal disclaimer. The court concluded its double patenting discussion with the following paragraph:
[947]*947As noted in Siu, extension of monopoly is not the only objection to double patenting. [fn. Underwood v. Gerber, 149 U.S. 224 [13 S.Ct. 854, 37 L.Ed. 710] (1893).] Others include possible harassment by multiple assignees, inconvenience to the Patent Office, and the possibility that one might avoid the effect of file wrapper estoppel by filing a second application, [fn. quoting from an H. Marans article, 36 JPOS 207 (1954), part of the discussion referred to.] We do not minimize those possibilities, but we must decide this case on the facts before us. We are not here confronted by a situation where any abuse of the terminal disclaimer is suggested. We conclude that on the facts here, the only real objection to granting appellant’s application is an extension of the monopoly. The terminal disclaimer, which Congress has expressly provided, removes any danger of such result, thus we are obliged to reverse the rejection of claims 7, 8, 9, 13 to 15 and 30 [for double patenting].
Thus, the court decided, on the facts of the case, to give no weight to the harassment possibility through multiple assignees, there being none, and that is the first item of case “law” on which appellants here rely. As a decision, the most the case stands for is that in the absence of any rule to the contrary, such as we now have in § 1.321(b), a terminal disclaimer without the transfer restriction now required was found acceptable. It did not hold that such a restriction would be improper.
In re Jentoft, like Robeson, was an obviousness-type double patenting case where applicant’s claims were rejected on claims of his own patent. He then filed a terminal disclaimer, citing Robeson as reason for accepting it. The examiner and board tried to distinguish Robeson on the ground that Jentoft’s patent contained a claim generic to the structure of the patent and to the modified structure of the application which was rejected on it. Conceding the terminal disclaimer would overcome the extension of monopoly objection to double patenting, emphasis was placed on the inability of the disclaimer to prevent the possibility of harassment in case the patent and any additional patent issuing on the application should fall into different hands because both patents would contain claims covering the apparatus of the application. It was contended that this could not have happened in the Robeson fact situation. In discussing that harassment issue, we began by saying:
As to the mere possibility of harassment, while we do not regard it as an impossibility, we think that giving weight to it to deny effect to terminal disclaimers is to overlook the countervailing advantages to the public, pointed out in Braithwaite * * *.
Braithwaite referred to two cases decided the previous year, 54 CCPA 1589 and 1604, 379 F.2d 594 and 606, 154 USPQ 29 and 38 (1967). There the court had pointed out that giving effect to terminal disclaimers— which the then Patent Office had refused to do — was in the public interest because it encouraged the disclosure of additional developments, the earlier filing of applications, and the earlier expiration of patents whereby the inventions covered became freely available to the public. In Jentoft we were still dealing with that much discussed question — were terminal disclaimers effective at all. The Patent Office was insisting that they were not in the Jentoft fact situation and pressing its argument on the basis of harassment possibilities. The court rejected the argument with some fervor. Looking at the Jentoft facts, the court thought the possibility of divided ownership “most unlikely.” The opinion refers to the “unreality of the harassment theory as a ground of refusing effect to terminal disclaimers.” But as the solicitor’s brief points out, what the court there thought unlikely has here actually occurred. Our earlier speculations must give way to reality.
What appellants have referred to as case law in conflict with § 1.321(b) consists of comments made in only two cases out of many in a rapidly developing field of jurisprudence fourteen or more years ago. Subsequent to those cases, it was the judgment of the administrative agency charged with [948]*948implementing the patent statutes, which is better informed on conditions in the field than are we, that the kind of terminal disclaimer commitment now contained in the regulation, referred to by this court in Griswold as an ingenious solution to a long-recognized objection to double patenting is desirable. It was first established as an administrative practice on Feb. 14,1968, by a Commissioner’s Notice, 848 O.G. 1, which introduced the non-alienability proviso. The Notice read:
The practice set forth in the notice of January 31, 1967, entitled “Double Patenting” (834 O.G. 1615), is modified to the extent that when a single inventive entity is involved, a terminal disclaimer will be accepted to avoid a double patenting rejection even if the claims overlap, if the claims which would otherwise be subject to such rejection could not have been allowed in the other application or patent, and if the terminal disclaimer further provides that the patent shall expire immediately if it ceases to be commonly owned with the other application or patent. [Our emphasis.]
This proviso was continued in a superseding notice of February 18, 1969 (860 O.G. 661). In 1970, the Commissioner published his Notice of Proposed Rule Making with a somewhat different provision and after a hearing and receipt of comments the present regulation was adopted in 1971 with the approval of the bar. It has now been in effect for more than a decade during which the bar has known what the procedure is. We have been shown no compelling reason to invalidate it.
Certainly many, if not most, double patenting situations fall into the obviousness-type double patenting category and involve a modification of or improvement upon what an inventor or his assignee has already patented. The desire is to be able to bring such improvement inventions within the protection of the patent system, at the same time giving an incentive for their disclosure. For a long time the judge-made law of double patenting was a serious obstacle to doing so. Knowing this, the drafters of the 1952 Patent Act provided a possible remedy in the terminal disclaimer, 35 U.S.C. § 253. See P. J. Federico, Commentary on the New Patent Act, 35 USCA p. 49 (1954). That provision is merely permissive and it was left to the courts to work out its application on a case-by-case basis. This court in the first Braithwaite case, 54 CCPA at 1598, 379 F.2d at 601, 154 USPQ at 34, speaking of such inventions and the granting of a second patent upon the filing of a terminal disclaimer making the two patents expire together, said:
When a terminal disclaimer causes two patents to expire together[,] a situation is created which is tantamount for all practical purposes to having all the claims in one patent.
Obviously, that thought contemplates common ownership of the two patents, which remains common throughout the life of the patents. In the discussions in the Robeson and Jentoft cases on which appellants rely, we gave little weight to the harassment possibility because we thought that divided ownership was most unlikely — a remote possibility. When it is a reality, as in this case, the situation is like that held impermissible in Pope v. Gormully, 144 U.S. 248, 12 S.Ct. 641, 36 L.Ed. 423 (1892), cited by the solicitor, where a patentee undertook to assign separate claims of the same patent to different parties.
Upon this extensive review of the situation, we consider it desirable to tie both the termination and the ownership of the two patents together, as is required by § 1.321(b), and, seeing no substantial obstacle to doing so, hold it to be a valid regulation.
The decision of the board is affirmed.
AFFIRMED.