Reid Harvey appeals from decisions of the Board of Patent Appeals and Interferences (Board) of the Patent and Trademark Office (PTO), which affirmed final rejections of Harvey’s design patent applications, Serial Nos. 07/172,904 (’904) and 07/172,906 (’906). The ornamental designs for vases of the ’904 and ’906 applications were held obvious under 35 U.S.C. § 103 (1988). Because we conclude that the Board misapplied the. obviousness standard to the application designs, misconstrued the combined teachings and combinability of the prior art references, and clearly erred in finding substantially identical visual appearance between the claimed designs and the .prior art designs, we'reverse.
BACKGROUND
The ’904 and ’906 applications éach disclose a vase formed by the intersection of two geometric solids. The ’904 application depicts a vase formed by the intersection of an “oblate ellipsoid with a pentagonal cylinder.” See Exhibit A attached. The examiner rejected this design application as representing an obvious modification of a prior art vase by Hárvey himself in view of a secondary reference to a bowl design, designated “shape 1982” in the publication “The Glass of Fredrick Carder,” copyrighted by P.V. Gardner in 1971 (Carder publication).
The Harvey prior art vase has a shape formed by the intersection of a sphere with a cube (see Exhibit B attached), and the Carder shape 1982 bowl has a shape which when rotated about its central axis forms an oblate ellipsoid (see Exhibit C attached, which shows a cross-section of the vase). Although the examiner conceded that the combined references did not disclose the exact shape of the claimed design, she contended that the claimed design differed from the references only in the specific pair of geometric solids used
(ie.,
ellipsoid versus sphere and pentagonal cylinder versus cube). According to the examiner, the substitution of one common geometric solid for another would have been
per se
obvious to one of ordinary skill in the vase design art. She cited
In re Hopkins,
The ’906 application depicts a vase formed by intersecting an “egg-shaped revolved hyperboloid with a truncated pyramid (also referred to as a square frustrum).” See Exhibit D attached. The examiner rejected this design as an obvious modification of the same prior art Harvey vase in view of “shape 2794,” a vase, in the Carder publication. See, respectively, Exhibits B and E attached. As regarding the ’904 application, the examiner conceded that the combined references did not disclose the exact shape of the claimed design. Nevertheless, the examiner found that the Carder shape 2794 was substantially identical to the shape' of the ’906 design, except for the upper lip, and that a truncated pyramid was ah obvious variation of a cube.
Both rejections were affirmed by the Board. The Board held that the prior art Harvey vase qualified as a “basic design” because it could be analyzed as an intersection of two solids, one a rotational solid and the other a cubical solid. Respecting the ’904 application, the' Board also found that the substitution of one rotational solid (an ellipsoid) for another (a sphere) was a routine matter in the vase or container art, and' that changing a sphere to an ellipsoid involved merely a change of scale on a single axis. The Board further found that the “visual impact of the difference between a ‘pentagonal cylinder’ and a cubical structure is *1063 minimal.” Ex parte Reid Harvey, Appeal No. 92-2395, slip op. at 4 (Aug. 5, 1992). Respecting the ’906 application, the Board likewise found that substitution of the Carder shape 2794 for the sphere was within the skill of the art and that a square frustrum or truncated pyramid had an essentially identical visual appearance to a cubical structure.' With respect to the fact that the ’906 design has no upper lip while the prior art Carder 2794 vase does, the Board stated that this difference was insignificant because “the utilization of [said] lip at the top of a vase, or its omission, represents an obvious variation well-known to designers of vases.” Ex parte Reid Harvey, Appeal No. 92-2394, slip op. at 4 (Aug. 5, 1992).
Although Harvey argued that numerous and extensive other changes would have to be made to the pair of designs in the cited prior art references in order to achieve each claimed design, the Board found that the differences involved only minute details, which did not significantly impact the overall appearance of the design. In support of its position, the Board cited
In re Rosen,
ISSUES
The issues confronting us are (1) whether the Board misapplied the standard of obviousness to these ornamental designs, (2) whether, even if it properly applied the standard, the Board nevertheless erred in holding the designs of the ’904 and ’906 applications obvious in light of the prior art’ Harvey vase and the respective Carder references, and (3) whether the Board’s underlying factual findings that differences between the prior art designs and each claimed design were minimal were clearly erroneous.
STANDARD OF REVIEW
Obviousness is a question of law which we review
de novo. Gardner v. TEC Sys., Inc.,
•ANALYSIS
I. Obviousness Standard Misapplied to the Designs
In ornamental design cases, a proper obviousness rejection based on a combination of references requires that the visual ornamental features (design characteristics) of the claimed design appear in the prior art in a manner which suggests such features as used in the claimed design.
In re Sung Nam Cho,
As the Board recognized, a primary reference (basic. design) must be cited having design characteristics which are “basically the same” as the claimed design.
In re Rosen,
On appeal, Harvey argues that the Board did not correctly apply the standard of obviousness because the Board improperly analyzed Harvey’s prior art vase as a “design concept” rather, than the disclosure of specific design characteristics, and-that this error of law led to an improper conclusion of obvi *1064 ousness. Harvey conceded that all three pri- or art designs originate from the same basic design concept, namely the formation of a vase or bowl by the intersection of two solids — a rotational solid with a cubical solid. However, according to Harvey, there was no suggestion to combine the prior art references. See discussion in section II infra.
Like the examiner, the Board improperly mixed principles of obviousness for utility patents with those for ornamental design patents. Unlike an invention in a utility patent, a patented ornamental design has no use other than its'visual appearance,
In re Glavas,
Furthermore, the facts in In re Hopkins, cited by the examiner, are distinguishable from those present here. First, the field of loud speaker design is much less crowded than that of ornamental vase design. Thus, the fact that a design patent for an octagonal outline loud speaker precludes issuance of a design patent for a circular outline loud speaker does not necessarily imply that a prior art vase formed by the intersection of a sphere and a cube precludes the later issuance of a patent for a vase design formed by the intersection of an egg-shaped hyperboloid with a truncated pyramid or an oblate ellipsoid with a pentagonal cylinder. Second, In re Hopkins involves the substitution of a single, simple, two-dimensional, geometric shape (an octagon) for another (a circle), whereas the present case involves the replacement of two, more complicated, three-dimensional figures (a sphere and a cube) with two highly complicated solids (an egg-shaped hyperboloid and a truncated pyramid or an oblate ellipsoid and a pentagonal cylinder). Third, the court in In re Hopkins did not purport to extend its holding beyond the facts before it, and neither should this court. Fourth, this 1929 opinion, which predates the 1952 Act and talks of a concept (“inventive in character”) now irrelevant, is at best, a weak, old and questionable precedent.
Because the Board erroneously applied the standard of obviousness test, and because Hopkins can be readily distinguished on the facts, reversal is appropriate.
II. Erroneous Conclusion of Obviousness Based on Evidence Misconstrued
Even if, however, the Board had correctly applied the standard of obviousness, its conclusion of obviousness would still be erroneous because the Board misconstrued the combined teachings and combinability of the prior art, and improperly compared the visual impressions of selected features, rather than the visual impression of the work as a whole.
First, the Board misfound the combined suggestions of the prior art, reading far more into it than warranted. As a result .of over-inflating the reference value of the prior art, the Board necessarily, although only impliedly, concluded that any rotational solid intersected with any linear solid would be obvious, at least if a picture of each geometric solid could be found in the prior art, and perhaps even if not. Clearly, such a conclusion would be inconsistent with our case law on design patents.
Second, the Board erred in its determination of obviousness because there was no suggestion in the prior art to combine the references and their visual ornamental features (design characteristics). The CCPA
*1065
held in
In re Glavas,
If we adopted the logic of the Board and concluded that the substitution of the Carder shapes for those in the Harvey prior art vase would render the ’904 and ’906 design applications obvious just because the Carder shapes were well-known and frequently used in vase designs, each and every prior art bowl or vase shape ever publicly disclosed would render obvious any. generally similar vase shape. Clearly, this cannot be the case.
Third, the Board erred in its determination of obviousness because it improperly compared the visual impressions of selected, separate features of the prior art designs to the ’904 and ’906 designs, rather than the visual impression of the designs as a whole. In
In re Rosen,
[i]n considering patentability of a proposed design the appearance of the design must be viewed as a whole, as shown by the drawing, or drawings, and compared with something in existence — not with something that might be brought into existence by selecting individual features from prior art and combining them, particularly where combining them would require modification of every individual feature, * * *.
(Emphasis added).
Applying the foregoing principles, we held in
Petersen Manufacturing Co. v. Central Purchasing, Inc.,
III. Clearly Erroneous Findings of Fact
Even if there were a suggestion to combine, it would still be necessary to make several significant changes in the cited prior art designs in order to achieve each of the claimed designs. In the case of the ’904 application, such material modifications include: (1) creating discontinuities at the top and bottom of the Carder 1982 shape to create an ellipsoid element and to provide for a top and bottom (the top of the claimed vase is not flat like the 1982 shape), and (2) changing the cube of the prior Harvey vase to a pentagonal cylinder. Similarly, in the case of the ’906 application, it would be necessary to: (1) eliminate the lip from the Carder 2794 vase, (2) change the shape of the 2794 vase by making it more egg-shaped rather than ginger-jar-shaped, and (3) change the cube of the prior Harvey vase to a truncated pyramid.
The court in
In re Carter,
*1066 Such was not the case with the Harvey-prior art design patent and the Carder references. It cannot fairly be said, as the Board did, that the differences between the pair of prior art references and each claimed design are “minimal,” or “so minor that they have an insignificant impact on the overall visual impression.” The ellipsoid of the ’904 design does not look like a sphere as in the prior art Harvey vase, nor like a flattened ellipse as in the Carder 1982 reference. The egg-shaped cylinder of the ’906 design does not resemble the modified ginger jar with a lip as in the Carder 2794 reference, nor does the pentagonal cylinder of the ’904 design resemble a cube. Accordingly, we must conclude that the Board’s findings that the differences between the prior art and Harvey’s applications were minimal, the necessary modifications de minimis, and the impact on overall visual impression minor, were all clearly erroneous.
IV: Factually Analogous Precedents
Finally, our conclusion of nonobviousness in design patent cases finds support in our precedents. In several cases with facts similar to those present here, we held that the prior art did not render the claimed design obvious., For example, in
Litton Systems, Inc. v. Whirlpool Corp.,
SUMMARY AND CONCLUSION
First, the Board erroneously applied the obviousness standard to the ornamental designs at- issue by relying upon the prior art Harvey vase as a design concept rather than as the disclosure of certain visual ornamental features. We, therefore, reverse on this basis. Second, even if the Board had correctly applied the standard of obviousness, reversal would still be necessary because there was no suggestion to combine the prior art visual ornamental features as Harvey did. Moreover, the prior art references required numerous material changes in order to achieve each claimed design, and none of these modifications were suggested in the prior art designs. Also, the modifications made in creating each claimed design resulted in it having a different overall visual impression. Third, the Board’s factual findings that the visual differences between the prior art designs and the claimed designs are minimal, de minimis or minor were clearly erroneous. Accordingly,' we reverse for these reasbns as well.
REVERSED.
*1067 Exhibit A '904 Harvey Application
[[Image here]]
Exhibit B Prior Art Harvey Design
[[Image here]]
*1068 Exhibit C Carder 1982 Design (cross section)
[[Image here]]
Exhibit D '906 Harvey Application
FIG. 2 FIG. 3 FIG. 4
[[Image here]]
[[Image here]]
Notes
. Although the Board did not actually cite In re Carter in its opinions, it did use the phrase "de minimis " in one of its opinions.