The July 2, 1987, decision of the Patent and Trademark Office Trademark Trial and Appeal Board (board) dismissing opposition No. 72,022 to the registration of the mark shown below as a trademark for "men's, women's and children's clothing--namely, blouses, pants, jackets, dresses, shirts, undergarments, socks and footwear," application Serial No. 493,403, filed August 6, 1984, claiming first use on July 5, 1984, is affirmed.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

Opposer, The B.V.D. Licensing Corporation, relies on its ownership and use of its unquestionably famous trademark B.V.D. or BVD, many times registered in various forms for the same class of goods as those named in the opposed application to register. Fourteen registrations issued from 1906 to 1983 are relied on herein. There is no priority issue as B.V.D. has been in use since 1876 and registered under the Trademark Act of 1905 since 1906, as evidenced by the many registrations of record.

The sole issue, as stated by the board, is whether opposer's and applicant's marks so resemble one another, the goods being at least in large part identical, that confusion is likely. 15 U.S.C. Sec. 1052(d). The board thought not. We agree with that conclusion though not with most of the board's reasoning in arriving at it.

The board proceeded on the assumption that opposer had to prove the fame of its mark by evidence in this proceeding. BVD, in our view, is, to most Americans at least, among which we count ourselves, practically a household word synonymous, primarily, with underwear for men. Opposer's registrations show it has been extended, during the past couple of decades, to the opposite sex. Webster's Third New International Dictionary (1971) contains the entry "B.V.D. * * * trademark--used for underwear." The Random House Dictionary of the English Language (1967) has this item: "B.V.D. Trademark. a suit of men's underwear, esp. a pair of undershorts. Also, BVDs. Cf. skivvy." Under "skivvy" is a cross-reference to "B.V.D." When a trademark attains dictionary recognition as a part of the language, we take it to be reasonably famous. (Webster has similar listings for "Kodak" and "Levi's.") See Coca Cola Co. v. Gemini Rising, Inc., 346 F. Supp. 1183, 1187 n. 1, 175 U.S.P.Q. (BNA) 56, 58 n. 1 (E.D.N.Y.1972), referring to a statement in The Business Lawyer of November 1971 that "B.V.D.," along with "Coca Cola" and "Singer," is one of "the three most-recognized trademarks in the world." The focus here, however, is not primarily on fame but on likelihood of confusion. Fame will be assumed. Courts may take judicial notice of facts of universal notoriety, which need not be proved, and of whatever is generally known within their jurisdictions. Brown v. Piper, 91 U.S. (1 Otto) 37, 42, 23 L. Ed. 200 (1875). To that end, dictionaries and encyclopedias may be consulted. United States v. Merck & Co., 8 Ct.Cust.Appls. 171 (1917). Fed.R.Evid. 201(b) and (f). We judicially notice the fact that within our jurisdiction, which is the whole United States, the B.V.D. trademark is at least widely, if not universally, known.

BVD or B.V.D. has arguable similarities to B A D that are too obvious for discussion. The practical question is whether they are significant. It also has differences which we will discuss. We give no weight to the argument of desperation that an "A" when inverted bears some resemblance to a "V". The purchasing public, we believe, does not indulge in such recognitional contortions but sees things as they are. It will not read an A for a V and, seeing an A between a B and a D will react to it as the common word "bad," not as a copy or simulation or suggestion of the well-known mark B.V.D. If that recognition makes no sense in the context in which the mark is observed, the viewer, if of the right generation, might think it was intended to mean good, as argued by applicant, a dictionary-recognized slang use of which we were not previously aware. On the other hand, seeing the periods, we believe it would be a normal and therefore common reaction to the symbol to think the letters must be the initials for some commercial entity other than opposer. Considering the norms of the commercial uses of company initials, we cannot imagine that applicant, Body Action Design, Inc., would put its mark into use without in some way associating it with what it is intended to convey, namely, the origin of the goods in that company whose initials constitute the major portion of the mark (all but the raised periods).

The fame of a mark cuts both ways with respect to likelihood of confusion. The better known it is, the more readily the public becomes aware of even a small difference. BVD has that well-known quality which would trigger the observer to notice at once that B A D, with or without the periods in either mark, is a different symbol. Consider the similar ruling, dismissing an opposition, by our predecessor court in Gulf States Paper Corp. v. Crown Zellerbach Corp., 417 F.2d 795, 57 CCPA 720, 163 U.S.P.Q. (BNA) 589 (1969), where the owner of the mark "E-Z", which is recognizable as the word "easy" in meaning and sound, failed to prevail in the attempt to prevent registration of "CZ". The court held the marks are not confusingly similar, one judge on a 5-judge court dissenting. The discerning, of course, could see that "CZ" stood for Crown Zellerbach.

For the above reasons, we agree that the opposition was properly dismissed. The decision of the board is

AFFIRMED.

NICHOLS, Senior Circuit Judge, concurring.

I join in the court's opinion. In view of the use made therein of the concept of judicial notice, I am particularly glad to see it. Judicial notice gets quite a workout in TTAB appeals, but it is not acknowledged as often as it should be. With so little at stake in the grant or refusal of a trademark or service mark registration, it is often not worthwhile to fill the record with proofs of every fact a court might wish to take into account. Judicial notice fills the gaps. It is just as well for a judicial author to say so when he is doing this, and every now and then to take a side glance at Fed.R. of Evid. 201 to be sure he is using judicial notice correctly. It is available at the appellate level, even if not used by the trial tribunal, and may support or undermine that tribunal's conclusions. Rule 201(f). It is particularly needed when the issue is "likelihood of confusion." To what extent can we resort to our "gut feeling" to supply evidence of what consumers will think falsely that is not in the record? By the Advisory Committee's notes, it must be "beyond reasonable controversy." We cannot use a debatable supposed fact to bootstrap ourselves to a dubious conclusion.

Amalgamated Bank of New York v. Amalgamated Trust & Savings Bank, 842 F.2d 1270 (Fed.Cir.1988), by another panel than this one, illustrates what I consider the proper use and misuse of judicial notice. We invoked by judicial notice what we supposed we, and everyone else, knew about how the banking business is conducted in this Year of Our Lord 1988. We professed not to know how "consumers," i.e., customers of banks get to be customers, as well as bankers do, so we would not substitute a contrary view for that of bankers, that identity of name does not by itself produce confusion. Actually, around the corner from our courthouse, on the same block of 15th Street, N.W., are two large full-service banks, independent of one another, but both using the word "American" in their names. If that produced customer confusion, it would seem they would do something about it. The temptation to point this out was resisted, as it does seem to be pushing judicial notice too far to draw conclusions from what might be an atypical situation. What if, however, there had been nothing else to go on? Arguably, if the parties on both sides fail to offer evidence on a relevant point, they stipulate by implication that the court will decide it by judicial notice.

It would be proper to take judicial notice that a lot of deregulation of banks had occurred in recent years, and that some of it has made a lot of difference to customers, for example the availability of interest on checking accounts. It would be all wrong to determine that deregulation has put customers' bank deposits at a greater risk of loss. It was all wrong for the board to say that deregulation had effected such changes that experienced absence of customer confusion in the past had no bearing on whether such confusion would be experienced in the future. In the absence of any showing of any item of deregulation that could have a bearing on confusion as to a bank's identity, this was a controversial inference, and because controversial, a misuse of judicial notice.

The importance of Rule 201 cannot be underestimated in light of the numerous cases in which questions of appropriate use of judicial notice arise. I am pleased with the panel opinion's reference to so undermentioned a rule.

NIES, Circuit Judge, dissenting.

I dissent. If the trademark sought to be registered were the word "BAD", I would agree with the majority's affirmance. However, the mark is not the word "BAD" nor is it displayed as a word. The mark is comprised of the initials of applicant's corporate name and is displayed B.A.D, that is, the letters are separated by periods or dots.

The record in this case amounts to little more than the application file and the numerous registrations of opposer for the mark B.V.D. Nevertheless, the record is sufficient to prove opposer has long-established, prior rights in B.V.D. for essentially the same goods sold under applicant's B.A.D mark. I agree with the majority that judicial notice may be taken that B.V.D. is a famous mark for underwear.

Applying the principles established by our precedent leads me to conclude that the prior registrations for the B.V.D. mark preclude registration of B.A.D because there is a likelihood of confusion between the marks as used for identical or closely related goods. Lanham Act Sec. 2(d), 15 U.S.C. Sec. 1052(d) (1982).

It is a basic principle that marks must be compared in their entireties for purposes of determining likelihood of confusion, not dissected into their component parts. See, e.g., Joseph Schlitz Brewing Co. v. Houston Ice & Brewing Co., 250 U.S. 28, 29, 39 S. Ct. 401, 401, 63 L. Ed. 822 (1919) (Holmes, J.) ("It is a fallacy to break the fagot stick by stick."); Specialty Brands, Inc. v. Coffee Bean Distribs., Inc., 748 F.2d 669, 672, 223 U.S.P.Q. (BNA) 1281, 1282-83 (Fed.Cir.1984). The reason for this principle is that the average purchaser, at least of low-cost shelf goods such as those involved here, is unlikely to perceive specific details of a mark. Colgate-Palmolive Co. v. Purex Corp., 451 F.2d 1401, 1402, 59 CCPA 741, 742, 172 U.S.P.Q. (BNA) 176, 176 (1971). Accordingly, our precedent has noted that purchasers have particular difficulty in distinguishing between combinations of letters. See, e.g., Crystal Corp. v. Manhattan Chem. Mfg. Co., 75 F.2d 506, 508, 22 CCPA 1027, 1029, 25 U.S.P.Q. (BNA) 5, 6 (1935); 2 J. McCarthy, Trademarks & Unfair Competition Sec. 23:13, at 75 (2d ed. 1984). See generally Vitamin Corp. v. American Home Prods. Corp., 166 F.2d 203, 35 CCPA 952, 76 U.S.P.Q. (BNA) 611 (1948) (mark "VCA" confusingly similar to mark "I.V.C.").

Second, it is another well established principle that likelihood of confusion may not be judged upon a side-by-side comparison of the marks. See, e.g., Geigy Chem. Corp. v. Atlas Chem. Indus., Inc., 438 F.2d 1005, 1007, 58 CCPA 972, 974, 169 U.S.P.Q. (BNA) 39, 40 (1971); 2 J. McCarthy Sec. 23:17, at 98. I believe the majority does not follow that principle. The majority presumes that the purchasing public will actually see that B. A.D is not B.V.D. because B.V.D. is well-known. In reality, the opposite is true. What happens is that a purchaser is less likely to perceive differences from a famous mark. All that is needed is a suggestion of such mark to trigger a mental perception that it is the famous mark. Purchasers simply do not take the time to study the marks and see the differences. Nor are they expected to. See Specialty Brands, 748 F.2d at 675, 223 USPQ at 1284 (fame important to likelihood of confusion inquiry because public may exercise less care in purchasing product under a famous name); 1 J. Gilson, Trademark Protection & Practice Sec. 5.03, at 5-21 (1987) ("Consumers under actual product selection circumstances rarely analyze trademarks minutely or compare their various elements."); 2 J. McCarthy Sec. 23:15, at 81.

Third, applicant argues, and the majority agrees, that purchasers will pronounce B. A.D as the word "BAD". There is no evidence how the public articulates B. A.D. Indeed, no evidence exists indicating how applicant itself pronounces its mark, whether as initials or as a word. Thus, under our precedent, it must be presumed that the mark will be pronounced either as initials or as a word. See Barton Mfg. Co. v. Hercules Powder Co., 88 F.2d 708, 710, 24 CCPA 982, 984, 33 U.S.P.Q. (BNA) 105, 107 (1937) (no "correct" pronunciation of a mark). Moreover, the derivations of the marks are of no particular significance. See Aerojet-General Corp. v. Computer Learning & Sys. Corp., 170 U.S.P.Q. (BNA) 358, 362 (TTAB 1971) (fact that letter marks are acronyms derived from different words unimportant because average purchaser probably unaware of derivation); 1 J. Gilson Sec. 5.02, at 5-18.

Finally, in close cases, all doubts must be resolved in favor of the prior user. See Geigy Chem., 438 F.2d at 1008, 58 CCPA at 975, 169 U.S.P.Q. (BNA) at 40; 2 J. McCarthy Sec. 23:21, at 106. As we have stated and restated:

The law has clearly been well settled for a longer time than this court has been dealing with the problem to the effect that the field from which trademarks can be selected is unlimited, that there is therefore no excuse for even approaching the well-known trademark of a competitor, that to do so raises "but one inference--that of gaining advantage from the wide reputation established by appellant in the goods bearing its mark," and that all doubt as to whether confusion, mistake, or deception is likely is to be resolved against the newcomer, especially where the established mark is one which is famous and applied to an inexpensive product bought by all kinds of people without much care.

Planters Nut & Chocolate Co. v. Crown Nut Co., 305 F.2d 916, 924-25, 50 CCPA 1120, 1128-29, 134 U.S.P.Q. (BNA) 504, 511 (1962) (footnote omitted); Specialty Brands, 748 F.2d at 674, 676, 223 U.S.P.Q. (BNA) at 1284, 1285 (court "compelled" to resolve doubts in favor of famous, established mark against newcomer, quoting Planters Nut ). That guideline should be applied here to protect a famous mark from encroachment with the unfortunate consequence that long-established rights will be narrowed. Because of this decision permitting encroachment by the mark B. A. D, B.V.D. is in a weaker position to assert rights against other similar marks. Under the majority's analysis, B. E.D, B. I.D, B. O.D, and B. U.D for underwear are also registrable because they also might be perceived as words.

This case brings to mind the oft-quoted truism: "[F]ew would be stupid enough to make exact copies of another's mark or symbol. It has been well said that the most successful form of copying is to employ enough points of similarity to confuse the public with enough points of difference to confuse the courts." Baker v. Master Printers Union, 34 F. Supp. 808, 811, 47 U.S.P.Q. (BNA) 69, 72 (D.N.J.1940); Plough, Inc. v. Kreis Laboratories, 314 F.2d 635, 644 n. 4, 136 U.S.P.Q. (BNA) 560, 567 n. 4 (9th Cir.1963) (Pope, J., dissenting); Armstrong Cork Co. v. World Carpets, Inc., 76 F.R.D. 613, 614, 198 U.S.P.Q. (BNA) 526, 527 (N.D.Ga.1977). That appears to me to be the situation here.

Accordingly, I would reverse.