BALDWIN, Judge.
This is an appeal from the decision of the Board of Appeals,1 affirming the examiner’s rejection of the sole remaining claim in appellant’s application2 for “Mixing Yial Construction” as “unpatentable over” either Bujan3 or Lock-hart4 in view of Jensen5 and either parsons et al.6 or Umbdenstock7 under 35 U.S.C. § 103. Following the first decision by the Board of Appeals, appellant requested reconsideration, and the board rendered a second decision denying any error in the first and making no changes therein.
In the decision from which this appeal stems, the board made reference to an earlier decision it had rendered in appellant’s parent application:
While we agree with the appellant that the Board’s decision in appellant’s earlier application, Serial No. 752,907, filed August 4, 1958, Appeal No. 15-87, is not res judicata as to the present claim which recites the additional element of a silicone coating, we agree with the Examiner that said decision is equally applicable to- the instant claim to the extent of the common subject matter, i. e., without the silicone coating. We deem it unnecessary to rehearse the same herein. Reference is, accordingly, made to said decision.
In view of this, appellant urged in his brief to this court:
* * * that an erroneous application of the doctrine of res judicata induced the final rejection of Claim 6 by the Primary Examiner and a subsequent affirmation of such rejection by the Board of Appeals, without due regard for the Appellant’s arguments distinguishing the claimed invention from the references. Appellant is convinced that the Examiner believed that Claim 6 could not be allowed unless it could be distinguished from the combination of a reference showing a silicone coating and the decision of the Board of Appeals in the parent application, Serial No. 752,097, with respect to the sub-combination of elements contained in its claims. Accordingly, the Appellant’s parent application was, in effect, improperly used as a reference against the recitations in Claim 6 and, in fact, the issue of unobviousness was not considered apart from the influence of the prior decision of the Board of Appeals in Appeal ,No. 15-87.
The Board of Appeals was also distracted by its prior decision from an [580]*580unbiased determination on the issue of unobviousness * * *
Furthermore, this constituted an improper application of res judicata in spite of the Board’s attempt to avoid such a conclusion in its * * * denial * * *
However, in view of the facts in this case, we agree with the Solicitor that:
There is no issue of res judicata in this case. The examiner made no mention of res judicata. * * * The Board of Appeals held flatly that the earlier decision “is not res judicata”.
Inasmuch as res judicata has never been urged as a ground of rejection by the examiner, by the board, or by the Solicitor, we do not perceive the existence of any such issue merely because of an incorporation by reference of the reasons for a holding of obviousness as previously stated in a prior decision; and we will, therefore, direct our attention to the issue of obviousness under 35 U.S.C. § 103.
THE INVENTION
The subject matter of the invention may be briefly summarized from appellant’s specification as follows:
This invention relates to an improvement in a plural compartment mixing vial construction and, more particularly, relates to an improved center seal plug which may be placed between the compartments for temporarily isolating same from each other. * *
It has been previously suggested to provide a two compartment mixing vial with a construction between the compartments thereof defining a seat against which a natural or synthetic rubber plug was seated to thereby isolate the compartments from each other. One of the compartments is adapted to contain a solid material, such as a desiccated pharmaceutical product, while the other compartment is adapted to contain a liquid, such as an aqueous diluent or solvent for the solid material. A piston is snugly and slidably disposed at one end of the liquid-containing compartment and is arranged so that it can be manually forced into said compartment to pressurize the liquid therein and thereby exert sufficient hydraulic pressure on the plug to discharge it into the solids-containing compartment whereupon the liquid enters the compartment and is mixed with the solid material therein. Such vials are especially well adapted for use as packages for those pharmaceutical products which are used in solution form but are best stored in dry or solid form. In particular, such vials are useful for packaging parenteral formulations which are not stable for a prolonged period of time and, thus, must be formed immediately before use, usually by mixing the medicinal ingredients which are in a sterile powder form with a suitable sterile diluent which consists essentially of water. However, prior vial constructions of this type have not attained substantial commercial acceptance because an excessive amount of moisture is transmitted from the liquid-containing compartment into the solids-containing compartment and, since many sterile powder formulations are sensitive to water, the medicinal agents become unstable during prolonged storage, particularly when stored at elevated temperatures. * * *
* * * * * *
I have discovered, unexpectedly in view of the prior art, that the above-mentioned problems are satisfactorily solved where the center seal plug for temporarily isolating the compartments from each other is formed of butyl rubber and the center seal plug is coated with a thin film of a silicone. * * *
The invention is more particularly reflected by appellant’s drawing reproduced here and by claim 6, the only claim remaining in the application, the improved center seal plug having been [581]*581emphasized therein for simplicity of identification:
6. A packaged pharmaceutical product comprising in combination:
a substantially tubular glass body being closed at one end thereof and open at the other end thereof and having a cylindrical portion of reduced cross-sectional area spaced from both ends of the body, said body having a first chamber and a second chamber on opposite sides of said cylindrical portion, said second chamber being adjacent said one end of said body, said cylindrical portion being of constant diameter between the ends thereof and providing a passageway connecting said chambers and defining a seat therebetween, the internal wall of said seat being free of any coating;
a one-piece, solid and cylindrical plug of butyl rubber having a coating consisting of silicone, said plug being non-toxic and having an outside diameter slightly larger than the inside diameter of said seat so that said plug is slideably removably positioned snugly within said cylindrical portion of reduced cross-sectional area to isolate said chambers from each other and thereby minimize the moisture vapor transmission between said chambers;
a quantity of desiccated material in said second chamber and a quantity of aqueous solvent substantially filling said first chamber; and
substantially cylindrical piston means snugly disposed in and closing the other end of said body, said piston means being slideably movable axially inwardly of said body to act through said solvent against said plug for displacing said plug into said second chamber to permit mixing of said solvent and said desiccated material.
The sole issue is whether the board was correct in holding that, at the time of the invention, appellant’s claimed invention was obvious to one of ordinary skill in the art in view of the references. 35 U.S.C. § 103. For reasons developed hereinafter, we agree with appellant that the invention was not obvious.
THE REFERENCES
The disclosures of the reference patents are summarized below in the order in which they appear in the rejection.
Bujan is directed to an improved piston closure assembly for a two-compartment mixing vial, which vial is quite similar to that of appellant, as may be seen from Bujan’s Fig. I reproduced below. The material of which Bujan’s center seal plug is made, is not disclosed, the only reference to the plug being merely “center seal member 17.” However, we accept the board’s inference in its opinion in appellant’s parent application (since it was incorporated by reference in the opinion herein) that Bujan’s “seal 17 may be made of natural rubber.” Bujan discloses no lubricant or film of any kind on; his center seal member and makes no reference whatsoever to the problem of moisture transmission from [582]*582one compartment to the other; accordiuo-lv. there is likewise nothing disclosed with respect to the solution of such a moisture transmission problem.
Lockhart also discloses a very similar mixing vial, as may be seen in Fig. 7 reproduced above, the alleged improvement being an elongated, substantially cylindrical center gate plug with enlarged and laterally flared ends, or flanges, to self-center the plug and keep it in place in its. seat. As disclosed, the center gate plug “may be formed or molded from any suitable elastic material, such as a rubber composition which may have a relatively high silicone content,” the high silicone content providing a slippery surface. Lockhart teaches that the flanged ends of his improved center gate plug avoid the problem encountered if a loader should accidentally push a substantially cylindrical, slippery gate plug, as used in the prior art vials, beyond the desired seating position; normally, such a prior art gate plug would have a tendency to move further downward and drop into the bottom chamber, but the flared ends of Lock-hart’s improved center gate plug are alleged to provide automatic self-centering. Despite Lockhart’s discussion of a slippery, “high silicone content” rubber, there is no suggestion of any surface coating, much less a silicone coating, on the center gate plug. Moreover, Lock-hart’s sole enunciation of the moisture transmission problem merits repetition herein:
It is important to prevent in such devices creepage of appreciable amounts of moisture from the liquid chamber to the solids chamber. Hygroscopic medicinal solids may be destructively affected in storage of such devices by the amount of moisture which may creep past the gate plugs. In devices of this type where contact between the gate plugs and their seats are in very narrow zones or is of a line contact nature microscopic fissures which are inherently present in the surfaces of molded and/or blown glass and similar materials permit such creepage past elastic plug surfaces during an initial storage period. Eventually the compressed elastic plug material is set into these fissures to attain a secure seal but by the time that develops the extent of moisture transfer may be such as to make the [583]*583medicament unsuitable for use. The long surface contact between the flare-ended gate plug and its seat, whether of the crown type illustrated in Fig. 3, or of the flat type illustrated in Fig. 7, efficiently prevents such moisture transfer, particularly during the initial portion of the storage period when the damage usually develops.
Jensen discloses a vessel with a sterile closure for the separate storage of two substances, which can be brought together before use without breaking the sterile closure. Jensen’s somewhat different partition for dividing the vessel into two compartments may be best appreciated by considering a portion of the specification with reference to Figs. 3-6 reproduced herebelow:
* * * the partition is constituted by a rigid disc-shaped body 42, made e. g. of metal, and an elastic ring 44 surrounding said body, the cross-section of the ring being rounded, preferably circular. The ring 44 may consist of rubber, especially butyl rubber (GR-I) which is impervious to steam. The circumference of the disc-shaped body is suitably circular, so that a gap, having the same width everywhere, will be provided along the periphery, when the body is disposed co-axially in the housing 10, the section of which is circular. In said gap the ring 44 is so disposed that it is subject to a pressure sufficient to provide safe tightening between the body 42 and the wall of the housing. A groove 46 is preferably provided at the edge of the body 42, in which groove the ring 44 normally rests.
******
When the partition is to be broken, the arched part 16 of the cap 14 is forced downwardly like a diaphragm. Thereby a pressure is transferred through the rod 48 to the disc-shaped body 42, and as a consequence thereof the ring 44 will be unrolled between the wall of the housing 10 and the edge of the body 42, as illustrated in Figs. 4-6. In Fig. 4 the parts assume the same relative positions as in Fig. 3. In Fig. 5 the ring 44 has left the groove 46, and the uppermost part of the edge of the body 42 has been pressed slightly into the ring 44, thereby increasing the stress of same. The body 42 will then be in a position of labile equilibrium, which is exceeded by its subsequent movement, see Fig. 0 * * *
The patent to Parsons et al. (hereafter “Parsons”) discloses that the provision of a silicone coating on rubber, synthetic rubber, or plastic stoppers for pharmaceutical containers “renders the stoppers more impermeable to moisture, renders them less apt to absorb moisture in treatment, and prevents them from stick[584]*584ing,” thusly providing stoppers which are “satisfactory for use in automatic stoppering machines.” The following objects of the invention are listed in the specification:
* * * to provide a lubricant for these rubber stoppers so that their tendency to stick to each other is reduced; to provide a surface sealing substance on the rubber stoppers to render the rubber more inert towards the contents of the vials; to reduce the permeability of the rubber stoppers to moisture; * * * and to provide a lubricant which will allow the rubber stoppers to be more easily inserted and withdrawn.
However, it should be noted that the downward movement of the easily slidable silicone coated stopper appears, from the drawings, to be limited by an external flange thereon which bears against the lip of the pharmaceutical container into which the stopper is inserted, and that auxiliary means may also be used to maintain and seal the closure provided by the inserted stopper since the specification further states, “The rubber stopper may be then fastened into the glass container by an aluminum seal or other standard type of seal.”
Umbdenstock also relates to “a rubber stopper having improved properties which is adapted for use in machines for automatically and mechanically stoppering vials or bottles,” particularly for “packaging certain drugs and therapeutic materials in bottles or vials under sterile conditions.” It is disclosed that natural rubber, reclaimed rubber, or synthetic rubber stoppers, if coated with a thin film of a high boiling point rubber lubricant, such as a silicone oil, may be used in automatic machinery without adhering to one another or to parts of the machinery and without jamming of the machines and that such silicone coated stoppers are characterized by much greater ease of insertion into bottles and vials. There is no drawing present nor any discussion in the specification to indicate the particular manner by which the silicone coated stoppers are maintained in position after they are inserted. Neither Parsons nor Umbdenstock discuss the in situ sealing qualities of the silicone coated rubber stoppers in great detail since both references are primarily concerned with providing suitable stoppers for automatic stoppering machines.
THE REJECTION
The examiner in his answer rejected claim 6 “as being unpatentable over either Bujan or Lockhart in view of Jensen and Parsons,” presumably under 35 U.S.C. § 103. Although Umbdenstock was listed by the examiner as a reference relied on, it was not applied to the claims by the examiner as acknowledged by the board:
The cited patent to Umbdenstock, while not specifically applied to the claims, is described at page 2 as alternate to Parsons et al. for the disclosure of the use of silicone as a coating on rubber plugs to enhance the ease of insertion into or removal of the plugs from surgical containers or vials.
Moreover, the board’s only discussion of that reference in its opinion, other than being listed as a reference relied on, was that “Umbdenstock similarly uses silicone coatings on rubber stoppers for the same purpose” as Parsons. Hence, we conclude that the examiner and the board have, at most, treated Umbdenstock merely as an alternative to Parsons; accordingly, it will be treated with Parsons and will not be further mentioned.
In affirming the examiner’s rejection, the board quoted with approval a portion of the examiner’s Answer:
“The Bujan and Lockhart references show appellants’ specific vial shape and center plug arrangement which is conceded by appellant. The use of butyl rubber as the material used for the center plug is believed obvious in view of Jensen whose seal 44 is constructed of butyl rubber and which is stated to be impervious to steam. In regard to the feature of coating the center plug with silicone for use as a lubricant or moisture inhibiter, such a feature is notoriously old in the art. [585]*585Lockhart * * * discloses the employment of silicone mixed with the plug for lubricating purposes. It is believed obvious to coat silicone on a sealing plug rather than to incorporate silicone into the rubber if so desired. That this is conventional is shown by the patent to Parsons * * * ’
The decision of the board further stated:
We are of the opinion, after a thorough review of the appeal record, including the affidavits of record and the oral arguments at the hearing, that the rejection of the claim at bar as unpatentable over the references is free of reversible error.
******
* * * Jensen clearly teaches in the art the use of butyl rubber for its steam, i. e., moisture, impervious property * * * and * * * in view thereof it would have been obvious pri- or to appellant entering the field to make the center plug 17 of Bujan of butyl rubber so as to render it impervious to moisture, the very purpose for which appellant employs butyl rubber.
In the instant case appellant has added a silicone coating to render the plug easier to insert or remove. This, we agree with the Examiner, is taught by Parsons * * *
******
The use of a silicone oil coating for easing the insertion and removal of the center of the plug of Bujan, made of butyl rubber as taught by Jensen, would appear to be obvious even to the uninitiated having these references before them.
******
Appellant has, in effect, merely combined features old in the art for their known purpose without producing anything beyond the results inherently expected therefrom.
OPINION
We believe that the Board of Appeals has erred in its application of 35 U.S.C. § 103 notwithstanding the last above-quoted statement, which, even if correct, does not conclusively establish that the invention is obvious under the patent law. A patentable invention, within the ambit of 35 U.S.C. § 103, may result even if the inventor has, in effect, merely combined features, old in the art, for their known purpose, without producing anything beyond the results inherent in their use. Although we believe that appellant, here, has actually done more than this in making his combination, we also believe that a more proper, albeit not exclusive, inquiry in a case such as this is to look further as to the reasons for making the combination.
It should not be necessary for this court to point out that a patentable invention may lie in the discovery of the source of a problem even though the remedy may be obvious once the source of the problem is identified. This is part of the “subject matter as a whole” which should always be considered in determining the obviousness of an invention under 35 U.S.C. § 103. In re Antonson, 272 F.2d 948, 47 CCPA 740; In re Linnert, 309 F.2d 498, 50 CCPA 753. The court must be ever alert not to read obviousness into an invention on the basis of the applicant’s own statements; that is, we must view the prior art without reading into that art appellant’s teachings. In re Murray, 268 F.2d 226, 46 CCPA 905; In re Sporck, 301 F.2d 686, 49 CCPA 1039. The issue, then, is whether the teachings of the prior art would, in and of themselves and without the benefits of appellant’s disclosure, make the invention as a whole, obvious. In re Leonor, 395 F.2d 801, 55 CCPA 1198.
Applying these principles to the case at bar and by reference to appellant’s specification, we find a clear indication that he discovered the source of the problem:
The problem of providing a structure for temporarily isolating a compartment containing a solid pharmaceutical product from a compartment containing an aqueous solution to prevent water transference therebetween has long been recognized by workers in the art and great amounts of time, effort and [586]*586expense have been expended to solve same but without commercially acceptable results, insofar as I am aware, prior to the present invention. It was believed that the water passed through the microscopic cracks and crevices in the mating walls of the vial and the plug which are inherently present as a result of the manufacturing operation thereof. Accordingly, the art has attempted to solve this problem by using a more deformable rubber plug and applying an increased pressure thereon when seating same against the internal wall of the vial so that the plug will deform to more completely fill such cracks and crevices. The art has also increased the length of the plug and the seat portion of the vial in order to provide a longer surface of contact between the seat and the plug to minimize such moisture transmission. Neither of these expedients has successfully solved the afore-mentioned problem.
* * * * * *
Contrary to the opinion of the other workers in the art, I have discovered that a primary cause of moisture transmission through natural rubber center seal plugs is the permeability of the plugs themselves to passage of moisture. That is, a substantial amount of the water which moves from the water-containing to the solids-containing compartment passes through the natural rubber plug rather than between the plug and the wall of the vial and, thus, no matter how tight a seal is provided between the plug and the vial, a substantial amount of moisture will be transferred, which amount has, in the past, been sufficient to deleteriously affect the solid contents of the other compartment, particularly where such contents are solid pharmaceutical ingredients sensitive to moisture.
The specification reveals: (1) that the problem facing the industry was due to moisture transmission into the lower compartment of the mixing vial; (2) that the appellant discovered the cause of the moisture transmission to be the passage of moisture through, rather than around, the center plug; and (3) that appellant solved the problem by fabricating a center seal plug of butyl rubber with a silicone coating. The crux of the matter in this case is the discovery by appellant that passage through the center plug was a major cause of moisture transmission.
Jensen disclosed that butyl rubber is impervious to steam, but it does not necessarily follow, as the solicitor argues, that “Certainly, it would be obvious to the ordinary artisan to utilize butyl rubber to take advantage of its moisture resistance as taught by the prior art since moisture transfer was a recognized problem with center sealing plugs.” This reasoning merely begs the question. The fact that moisture transmission was a recognized problem does not establish that knowledge of moisture transmission through, rather than around, the plug, was likewise either recognized or obvious. The question here is whether the prior art recognized the cause of the problem.
There is no teaching in the prior art which would suggest the necessity of selecting a center seal plug material which is more impervious to liquid water than Bujan’s or Lockhart’s natural rubber. As we pointed out earlier, Bujan makes no mention of any moisture transmission problem, and Lockhart’s only discussion of the problem refers to moisture passage between the plug and the vial caused by “microscopic fissures inherently present in the surfaces of molded and/or blown glass and similar materials.” Jensen’s disclosure of the impermeability of butyl rubber to steam does not render appellant’s invention obvious since neither it nor any other reference suggests that greater impermeability to liquid water is desired or necessary in a center seal plug. Likewise, Parsons’ mention of moisture permeability does not appear to be directed to permeability to liquid water. That a natural rubber plug might be permeable to steam in no way establishes or makes obvious its permeability to liquid water; thus the cause of the problem is not suggested by the prior art.
[587]*587Notwithstanding our analysis above, concerning appellant’s recognition of the source of the problem and solution thereof, we believe that the multi-reference rejection affirmed below is improper for reasons existing within the disclosures of the references themselves, namely, that the references themselves teach away from the combination. First butyl rubber does not slide readily against a glass surface because of its frictional properties, as evidenced by the rolling operation of Jensen’s butyl rubber sealing ring as it disengages the metal disc (see Figs. 4-6 above). Both the operability and utility of Jensen’s vessel depend upon this frictionally induced rolling action; in contradistinction, a sliding engagement is absolutely essential to the operability of appellant’s center seal plug. Hence, Jensen’s center seal is totally incapable of serving appellant’s purposes. Moreover, since butyl rubber is significantly less resilient and more rigid than natural rubber, Lockhart’s flanged center gate plug, if made of butyl rubber, would be difficult, if not impossible, to either seat or displace from its seat. Thus, a combination of either Lockhart or Bujan with Jensen alone would produce a seemingly inoperative device. Still further, since butyl rubber is significantly less compressible and less resilient than natural rubber, one might conclude that a butyl rubber plug should be cylindrical and should be made to close tolerances with the seat; but then, if a silicone coating were applied as a lubricant one might also conclude that there would be nothing, like the flange of either Parsons or Lockhart, to prevent the plug from falling through.
In addition to these reasons for considering the invention to be unobvious, a variety of collateral factors,8 suggesting the same result, have been evidenced by appellant’s numerous affidavits. The commercial success of appellant’s device, as suggested by the affidavit of Gauss, is an important indicia of unobviousness, particularly where, as here, it is coupled with other affidavits which suggest that the two-compartment vial had not been a success prior to appellant’s invention. The Enz affidavit evidences the efforts and failures of other research specialists and other organizations to solve the problem of moisture transfer and of the long standing need for a commercially acceptable solution to the problem. In this latter regard, we are mindful of our decisions in In re Tierney, 388 F.2d 1018, 55 CCPA 884, and In re Moore, 390 F.2d 1003, 55 CCPA 944. In this situation, these collateral factors merely reinforce our belief that the invention was not obvious.
For the foregoing reasons, the decision of the Board of Appeals is reversed.
Reversed.
WORLEY, C. J., concurs in the result.
SMITH, J., participated in the hearing of this case but died before a decision was reached.