Concurring opinion filed by Circuit Judge MICHEL.
PER CURIAM.DECISION
Champagne Louis Roederer, S.A. (“Roe-derer”) appeals from a decision in an opposition proceeding by the Trademark Trial and Appeal Board of the Patent and Trademark Office (the “Board”). See Champagne Louis Roederer, S.A. v. Delicato Vineyards, Opposition No. 80,932 (TTAB June 25, 1997). Delicato Vineyards’ (“Delicato”) predecessor in interest filed application serial no. 73/701,-485 to register the word- mark “CRYSTAL CREEK” for wine. Roederer filed an opposition to the registration based on its two marks — the word mark “CRISTAL,” and the mark “CRISTAL CHAMPAGNE” with accompanying graphic design. The Board dismissed Roederer’s opposition after, determining that the application mark was not confusingly similar to either of Roederer’s marks. Because we .conclude that Roederer failed to demonstrate any error in the Board’s legal analysis or ultimate legal conclusion regarding likelihood of confusion, or clear error with respect to any of its findings on the individual DuPont factors, we must affirm.
DISCUSSION
We evaluate for correctness the Board’s ultimate legal conclusion as to the likelihood of confusion between the application and opposer’s marks, but we review the underlying factual findings of the Board for clear error. See Kenner Parker Toys v. Rose Art Indus., 963 F.2d 350, 352, 22 U.S.P.Q.2d 1453, 1455 (Fed.Cir.1992); Weiss Assocs. v. HRL Assocs., 902 F.2d 1546, 1547-48, 14 U.S.P.Q.2d 1840, 1841 (Fed.Cir.1990).
Although the opposition proceeding, and thus the record, was characterized by a lack of evidence on many of the DuPont factors, the Board-did, to the extent it was presented with evidence, assess the thirteen factors enumerated by one of our predecessor courts as constituting the test for likelihood of confusion. See Application of E.I. DuPont DeNemours & Co., 476 F.2d 1357, 1361, 177 U.S.P.Q. 563, 567 (CCPA 1973). Specifically, the Board found or presumed that the class of goods for which the marks were used were the same (wine, including champagne), that those goods traveled in the same trade channels, and that the goods were purchased by the same or similar customers. See Champagne Louis Roederer, slip op. at 8-9. Certainly there were no restrictions in the application that would support any contrary findings. The Board also found that Roederer’s marks had “come to serve as a very strong indication of origin for [its] champagne.” Id. at 9. The Board, however, treated the dissimilarity of the marks *1375with respect to appearance, sound, significance, and commercial impression as the dis-positive DuPont factor, concluding that this dissimilarity alone precluded any reasonable likelihood of confusion. See id. at 11. Specifically, as to significance, the Board found that the word marks “CRISTAL” and “CRYSTAL CREEK” evoked very different images in the minds of relevant consumers: while the former suggested the clarity of the wine within the bottle or the glass of which the bottle itself was made, the latter suggested “a very clear (and hence probably remote from civilization) creek or stream.” Id. at 12. The Board then found that the appearance and sound of the competing marks were also dissimilar. See id. at 13. Based on these three underlying findings on this single DuPont factor, the Board dismissed the opposition because of the difference in commercial impressions it found were created by the marks.
We must affirm, for Roederer on appeal merely restates its arguments from the opposition, which primarily concern the appropriate weight to be given the various findings on the several DuPont factors considered by the Board, and has failed to demonstrate any reversible error in the Board’s decision or foundational factual findings.
Roederer suggests, first, that it was an error of law for the Board to rely solely on the dissimilarity of the marks in evaluating the likelihood of confusion and to fail to give surpassing weight to the other DuPont factors, all of which were found to favor Roederer. We note, however, that we have previously upheld Board determinations that one DuPont factor may be dispositive in a likelihood of confusion analysis, especially when that single factor is the dissimilarity of the marks. See, e.g., Kellogg Co. v. Pack’em Enters., 951 F.2d 330, 332-33, 21 USPQ2d 1142, 1144-45 (Fed.Cir.1991) (stating that “[w]e know of no reason why, in a particular case, a single DuPont factor may not be dispositive” and holding that “substantial and undisputed differences” between two competing marks justified a conclusion of no likelihood of confusion on summary judgment); Keebler Co. v. Murray Bakery Prods., 866 F.2d 1386, 1388, 9 U.S.P.Q.2d 1736, 1739 (Fed.Cir.1989) (agreeing with the Board that the “more important fact for resolving the issue of likelihood of confusion ... is the dissimilarity in commercial impression between the marks”). We have not been persuaded that, on this record, the Board erred in concluding that the marks’ dissimilarities were dispositive, notwithstanding due weight being accorded to the DuPont factors found in Roederer’s favor.
Second, Roederer has failed to demonstrate any instance of clear error in the Board’s factual findings with respect to the dissimilarities of the marks in appearance, sound, significance, or overall commercial impression. Although Roederer disagrees, chiefly, with the Board’s interpretation of the commercial impression of the marks, even reasoned disagreement with such a finding does not, without more, establish that it is clearly erroneous. Although this opposition may have presented to the Board a close question with respect to the dissimilarities of the marks in sound, appearance, significance, and commercial impression, we remain unpersuaded that the Board made any clear error in its factual findings and thus we decline to disturb the Board’s decision on this basis either.
Thus, because Roederer has failed to demonstrate any reversible factual or legal error, we must
AFFIRM.