Opinion PER CURIAM.
PER CURIAM:These appeals question the validity of four regulations promulgated by the Treasury Department’s Bureau of Alcohol, Tobacco and Firearms in an effort to set standards for labeling and advertising grape wine. Intensive study of these rules satisfies us that one is well buttressed by the record, and accordingly we uphold it. The other three, we find, are bereft of support either in evidence or logic, and these we remand to the agency for further consideration.
I. BACKGROUND
A. The Regulatory Framework
At the close of the Prohibition Era, Congress set out to regulate the resurgent industry in alcoholic beverages. In 1935, Congress passed the Federal Alcohol Administration Act1 which, among other things, proscribes interstate commerce in wine unless it is ú
labeled in conformity with such regulations, to be prescribed by the Secretary of the Treasury ... as will prohibit deception of the consumer . . . and as will prohibit, irrespective of falsity, such *140statements ... as the Secretary of the Treasury finds to be likely to mislead the consumer; ... as will provide the consumer with adequate information as to the identity and quality of the products .. . [and] as will prohibit statements on the label that are ... false [or] misleading ----2
The Act also outlaws wine advertisements that are inconsistent with the labeling standards.3
In 1978, the Bureau, acting for the Secretary,4 amended a number of implementing regulations, including the four types at issue in this litigation.5 One of the contested rules involves varietal wines — those, such as Chardonnay or Pinot Noir, made from a particular variety of grape — -and permits a wine label to carry the name of a given grape-type “if the wine derives ... at least 51 percent of its volume from that variety of grape.”6 Thus a wine may, consistently with the regulation, be labeled “Chardonnay” so long as it owes 51 percent7 of its volume to Chardonnay grapes, even though the rest is contributed by grapes of some other kind.
The second group of pertinent regulations are what might be called geographical rules. They allow a wine to be represented as the product of a given area — such as “Napa Valley” or “Sonoma County” — when in fact the percentage of its grapes raised in the locality indicated may be as low as 75,8 or 85 if a so-called “viticultural area” is connoted.9 Beyond that, and without caveat to the consumer, a wine label may utilize a corporate or trade name that includes a geographical term although none of the wine comes from the area suggested.10 In contrast, a brand name containing words of geographical significance cannot be used on a wine label “unless the Director [of the Bureau] finds that such brand name, either when qualified by the word ‘brand’ or when not so qualified, conveys no erroneous impressions . ...”11
The regulations further permit a winery to state on a wine’s label that it “produced” the wine when it may have fermented and clarified only 75 percent of the bottle’s contents.12 A winery may also declare that it “made” a wine when actually the wine was produced elsewhere and then shipped to the “maker” merely for some cursory treatment.13
B. The Rulemaking Proceedings
The regulations under attack are identical or substantially similar to forerunners *141promulgated in 1935,14 which were reconsidered by the Bureau during the 1970’s. A brief review of those proceedings will put the case in helpful context.
The Bureau first undertook revision of the labeling requirements in 1975.15 The next year, in response to industry and consumer comments, it published two proposals relevant here. First, it suggested a prohibition on brand names signifying geography unless the word “ ‘brand’ is [employed in] direct conjunction therewith.” 16 The reason given was that “[p]atently, the use on labels of geographical names which bear no relationship to the origin of the grapes used to make the wine can be confusing to the consumer.”17 At the same time, the Bureau recommended modification of the rules governing representations as to the origin of wine grapes to insist that 95 percent of wines carrying a vintage date, and 75 percent of all other wines, come from the area indicated on the label.18
In 1977, the Bureau recommended even more sweeping revisions of the regulations. One would be an increase in the required percentage of varietal grapes from the current 51 to 75, and mandatory disclosure on the label of the proportion of such grapes.19 Another, in response to comments that “regulations relating to trade names . . . mislead the consumer as to the identity of the producer or maker,” would “amend[] regulations designed to remedy the misleading aspects of the section”20 by prohibiting altogether the use of “made by,” and limiting application of “produced by” to wineries that had actually fermented and clarified 95 percent of the wine so labeled.21 Still others, pertaining to geographical representations, would require 95 percent of the grapes to come from the area indicated on a label giving a vintage date or naming a vineyard, 85 percent to be produced at a stated point of origin when that point is a viticultural area, and 75 percent to be grown in a stated political subdivision represented as the source of a wine so labeled.22
The Bureau issued its regulations in final form in 1978, with major revisions, however, of the proposals. The 75 percent varietal minimum was retained, though the mandatory varietal percentage statement was dropped.23 Redefinition of winemaking expressions such as “made by” and “produced by” was also scrapped.24 The minimum geographical percentage was kept at 85 for wine labeled as originating in viticultural areas and 75 for wine labels giving the name of a political subdivision, but the 95 percent minimum for vintage-dated or vineyard wines was eliminated.25 Lastly, the proposed prohibition on misleading brand names, after minor modification, was incorporated.26
*142C. The District Court Proceedings
Three sophisticated wine consumers27 who participated in the proceedings leading to adoption of the new regulations challenged four of them in the District Court.28 Each of the four is invalid under the Federal Alcohol Administration Act, they argued, for each indulges wineries in statements not wholly accurate. It is deceptive, they said, for a winery to label a wine “Chardonnay” when in fact only 75 percent of the grapes from which it was made actually were of that type; for it to represent that a wine comes from Napa Valley when only 85 percent of its grapes were grown there; or for it to proclaim that it “made” a wine that in fact it purchased.29 Indeed, the consumers continued, a wine label that does not list by percentages each grape-type used and each area represented is misleading and inadequate as a matter of law.30
The District Court, after reviewing the administrative record, agreed in part. Deeming the statutory language regarding false and misleading statements “clear and direct,” the court was disposed to “rely to the extent possible on the ordinary meaning of the words used, as it is the understanding of the ordinary consumer which must control.”31 From this perspective, the court reasoned that “[t]he clear implication of a label bearing only ‘Chardonnay’ is that the wine is made from the Chardonnay variety grape and no other. This is the ordinary meaning even though the label does not use ‘all Chardonnay’ or ‘100% Chardonnay.’ ”32 Rejecting the justifications proffered by the Bureau for the regulations as framed, the court concluded:
There is thus no basis, in law or in fact, for permitting the use of a varietal name on wine labels without disclosing that the identified grape variety does not represent 100 percent of the content.. . .
[T]he percentage of the labeled variety must be truthfully communicated. The same rationale requires that wine labels making representations as to geographic origin or the nature of specific winemaking operations should couple such representations with a more precise indication of what minimum percentages or other meanings attach to the claims made.
Congress has directed that wine labeling terminology be understandable to the ordinary consumer. In the Court’s view, this at least requires wine labels to carry concise explanations of any terminology used where principal grape types, chief geographic origins, and the identity of a producer or maker is represented to the consumer. The explanations must at least specify applicable minimum percentage standards, and are to be suitably featured in the context of the label as a whole.33
On the other hand, the District Court refused to grant a request for an order mandating a more elaborate labeling statement, notwithstanding the statutory call for regulations requiring labels providing consumers with “adequate information as to the identity and quality of the products.” 34 Said the court,
[ujnlike the absolute prohibition against false or misleading statements, the separate and different obligation to ensure that disclosure is adequate squarely implicates discretionary judgment on the part of the Secretary .... [Tjhe statute itself offers no guidance as to what constitutes “adequacy,” nor is such a determination feasible by reference to the ordinary meaning of the words used. The question of whether a label, otherwise not misleading, provides adequate informa*143tion is not therefore a matter of statutory construction.35
The Bureau’s assessment on what information need be furnished consumers, the court found, was rational and supported by record evidence.36 The court accordingly awarded the consumers summary judgment “with respect to the regulation’s failure to require truthful and accurate representation of principal varietal designations and appellations of geographic origin, and truthful and accurate identification of wine producers and makers,”37 and granted the Bureau, summary judgment “on all other aspects of the regulation.”38 Both the consumers and the Department of the Treasury have appealed.39
II. ANALYSIS
When reviewing agency action under the “arbitrary and capricious” standard,40 the task here,41 “the court must consider whether the decision was based on a consideration of the relevant factors. . . . Although this inquiry into the facts is to be searching and careful, the ultimate standard of review is a narrow one. The court is not empowered to substitute its judgment for that of the agency.”42 To uphold the agency action, however, the court must find a “rational basis for the agency’s decisions ... in the facts of record,”43 and must ensure that the agency “has demonstrated a ‘rational connection between the facts found and the choice made.’ ”44 On reviews of this type, “we . . . reverse for a ‘clear error of judgment’ only if the error is so clear as to deprive the agency’s decision of a rational basis.”45 With these principles firmly in mind, we consider each of the regulations challenged.
A. Varietal Labeling
We sustain the Bureau’s varietal labeling rule.46 The word “misleading” is no more self-defining than the word “ade*144quate”47 — or, for that matter, other statutory terms such as “unfair or deceptive”48 and “false and deceptive,”49 which are to be given meaning, at least in the first instance, by agency interpretation.50 To be sure, the District Court’s reading of the governing statutory language51 is logical, and one that many would accept, but that does not necessarily mean that the agency’s construction is irrational. There is ample support in the record for the Bureau’s conclusion that varietal percentage listings
would add little by way of useful information to the consumer. Comments and testimony indicated that percentage requirements might mislead the consumer to believe that the higher the varietal percentage, the better the wine. Additionally requiring a percentage statement for this item would logically result in percentage and definitional statements being required for all other labeling designations. This would produce an extremely cluttered label, cause cost problems, and be of uncertain assistance to consumers. The Bureau believes that there has been an insufficient showing of a misleading, effect on the public to justify requiring all these percentages and definitions to appear on the label.52
Instead, a public information program planned by the agency seemed adequate for the time being.53 Indeed, the statutory language itself indicates that no simple test of truth can be used to determine what is misleading, for the Secretary is directed to prohibit “irrespective of falsity . . . statements .. . likely to mislead the consumer.” 54
In the end, then, the Bureau settled on 75 percent minimum varietal content as the best of the available alternatives.55 It explained:
Varietal labeling provides a more informative labeling designation than des*145ignations such as Chablis, Burgundy, or simply red table wine. Requiring 100 percent varietal content for these wines would discourage the use of these designations and not necessarily improve the product. Some varieties grown under certain soil and climatic conditions and with certain viticultural practices may make excellent wines at the 100-percent varietal requirement. However, more often some blending with other varieties is necessary to arrive at the best product. The amount of blending will vary among different varieties and different harvests. Because of these factors we believe a certain amount of flexibility is necessary.
We have determined, however, that the 51-percent requirement is too low for most varieties, and permits so much blending that the finished product may bear little or no resemblance to the grape variety used to designate the wine. Based upon all available evidence we have concluded that the 75-percent requirement allows sufficient opportunity for blending while maintaining the identity of the grape.56
We thus perceive the Bureau’s varietal labeling regulation as the product of a reasoned and amply elucidated process. That a different result may also be plausible does not make the administrative outcome arbitrary or irrational. We could uphold the District Court’s decision to set the regulation aside only were we to substitute our assessment for the agency’s judgment. That we are not at liberty to do.
B. Geographical Terms
The situation presented by regulations governing use of terms of geographical significance in wine labeling57 is more complicated. The controversy over percentage labeling in such instances is similar to that in connection with varietal representations, but the Bureau has pointed to virtually nothing in the record supporting its position that labels featuring geographical terminology unaccompanied by grape-origin percentages are not misleading. Furthermore, aside from a boilerplate statement that “[w]e believe these regulations will inform wine consumers of those things which are useful to them”58 — which appears in the preamble to the final set of regulations— the Bureau has not provided any rationale touching on the Act’s consumer-protection purposes for declining to exact minimum percentage labeling in such instances.59
The regulation involving brand names incorporating geographical language underscores the problem. At one point in the rulemaking process, the Bureau declared that “[pjatently, the use on labels of geographical names which bear no relationship to the origin of the grapes used to make the wine can be confusing to the consumer.”60 Nonetheless, the Bureau never explained why it chose to regulate misleading geographical brand names but not equally deceptive geographical corporate or trade names.61 We have not been referred to anything in the record indicating such disparity in treatment, nor has the agency supplied a rationale that even attempts to harmonize these regulations with the Act. We have no alternative, then, but to direct a remand of these regulations to the Bureau for further consideration.
C. Winemaking Terms
At one time, the Bureau proposed revisions “designed to remedy the possibl[y] misleading aspects of the” regulations *146defining such winemaking terms as “produced” and “made by.”62 But ultimately, and despite vigorous complaints by consumers,63 the agency decided to continue in force rules which allow wineries fermenting as little as 75 percent of a given wine to represent that they produced it,64 and which permit those contributing very little to the final product to say that they made it.65
These regulations cannot be upheld, at least at present. The Bureau has brought to our attention no record material to justify its action. The agency’s statutorily-required “concise general statement” of reason66 is much more concise than informative; in full, it merely declares that “[ajfter considering the comments and testimony, the Bureau has decided not to issue new regulations on the subject of winemaking terms at this time.”67 It is by no means intuitively clear why it is not misleading for a winery to represent that it produced a wine when another was heavily involved in its production, or that it made a wine that in fact it purchased. These regulations, along with those dealing with geographical terms, must be returned to the Bureau.
III. DISPOSITION
“In wine there is truth,” goes the proverb, and the Federal Alcohol Administration Act requires wine labeling and advertising to be truthful as well. While it certainly is the Bureau’s role to determine in the first instance what makes labels deceptive or not, it has not endeavored to explain — either by reference to the records or by a reasoned statement — just how three sets of the challenged rules will curb misleading representations by wineries. Thus, while we cannot endorse the District Court’s view that the Act itself mandates particularly-formulated statements on wine labels, the regulations addressing geographical and winemaking terms must be sent back to the Bureau to afford it an opportunity to show that they meaningfully control misleading labeling and advertising. If that cannot be done, these regulations will have to be rewritten in such fashion that the agency can demonstrate compliance with the statutory mandates.
Accordingly, the order in Nos. 80-1086 and 80-1133 is reversed insofar as it relates to varietal labeling and affirmed in all other respects. The order in No. 80-1244 is vacated and the appeal therefrom is dismissed. The litigation is remanded to the District Court with the direction that it remand to the Bureau the regulations pertaining to geographical and winemaking terminology for reconsideration in light of this opinion.
So ordered.
. 27 U.S.C. §§ 201 et seq. (1976). For the early history of governmental regulation in this area, see National Distributing Co. v. United States Treasury Dep't, 200 U.S.App.D.C. 133, 139-142, 626 F.2d 997, 1004-1006 (1980).
. 27 U.S.C. § 205(e) (1976).
. Id. § 205(f) (1976). For this reason, our analyses and conclusions, though focused primarily on the labeling rules, are equally applicable to regulations correspondingly governing wine advertising.
. The Secretary has delegated to the Bureau plenary authority to administer the Act. See 37 Fed.Reg. 11696 (1972).
. The District Court spoke of “three specific aspects of the final regulations” in controversy. Wawszkiewicz v. Department of Treasury, 480 F.Supp. 739, 742 (D.D.C.1979). As we understand the record, however, four species came under attack there, see Wawszkiewicz v. Department of Treasury, 480 F.Supp. 739 (D.D.C. 1979), Complaint at 5-11, Joint Appendix (J.App.) 8-14, and certainly four are challenged on appeal.
. 27 C.F.R. § 4.23 (1980). The percentage of varietal grapes required will rise to 75% in 1983. Id. § 4.23a(b).
. Or, from 1983 onward, 75%. See note 6 supra.
. 27 C.F.R. §§ 4.25(a), 4.25a(b) (1980).
. Id. § 4.25a(e)(3)(ii). Viticultural areas, which are to be delimited in future rulemaking proceedings, are “grape growing region[s] distinguishable by geographical features.” Id. § 4.25a(e)(l)(i).
. Id. § 4.33(b).
. Id.
. Id. § 4.35(a)(1).
. Id. § 4.35(a)(2). In the rather cryptic language of the regulation, a winery may say that it “made” a wine if it “made or treated the wine, otherwise than as described in paragraph (a)(1) of this section ....” The paragraph referred to is § 4.35(a)(1), which defines a “producer” as one who ferments and clarifies at least 75% of the wine labeled.
. See Wawszkiewicz v. Department of Treasury, supra note 5, 480 F.Supp. at 741-742.
. See 40 Fed.Reg. 30117 (1975).
. 41 Fed.Reg. 50004, 50005 (1976).
. Id.
. Id. at 50004.
. 42 Fed.Reg. 30517 (1977).
. Id.
. Id.
. Id. at 30518.
. 43 Fed.Reg. 37672 (1978).
. Id. at 37674.
. Id. at 37673.
. Id. at 37674. As originally suggested, the regulation would have permitted geographical brand names only if qualified by the word “brand” in type as “conspicuous as the brand name,” and trade and corporate names with geographical significance only if approved prior to the date of the proposal. 41 Fed.Reg. 50004, 50007 (1976). That was later modified to allow the word “brand” to appear in type smaller than the brand name. 42 Fed.Reg. 30518, 30521-30522 (1977). In final form, the regulations require use of the word “brand” only in conjunction with brand names found to be misleading, and the grandfather provision for trade and corporate names was eliminated entirely. 43 Fed.Reg. 37672, 37674 (1978) (“[t]he new regulations will not affect the use of corporate or trade names”).
. Throughout this opinion, we refer to these parties simply as consumers.
. Wawszkiewicz v. Department of the Treasury, supra note 5. See note 5 supra.
. See Wawskiewicz v. Department of the Treasury, supra note 5, 480 F.Supp. at 742.
.Id.
. Id. at 743 (citations omitted).
. Id. at 743-744.
. Id. at 745 (footnote omitted).
. See text supra at note 2.
. Wawszkiewicz v. Department of Treasury, supra note 5, 480 F.Supp. at 745-746.
. Id. at 746-747.
. Id. at 747.
. Id.
. The principal appeals, No. 80-1086, Department of Treasury v. Wawszkiewicz and No. 80-1133, Wawszkiewicz v. Department of Treasury, have been consolidated, id. (D.C.Cir. Mar. 11, 1980) (order), and with them a third-appeal, No. 80-1244, Wine Inst. v. Wawszkiewicz. The latter protests an order of the District Court denying a post-judgment motion by the Wine Institute, an industry group, to intervene. The Institute argues that the court effectively “prescrib[ed] the content of new rules” by directing that definitions and minimum percentages be listed on wine labels, thus “eliminat[ing] Wine Institute’s ability to participate in the formulation of those rules.” Reply Brief of Appellant Wine Institute at 2. The Institute concedes, however, that “[i]f the District Court had remanded the case to Treasury with instructions to reconsider the need for different rules ... Wine Institute could have participated in the rulemaking ...,” id. at 1, and thus in effect that intervention would then have been unnecessary. In light of our conclusion that the Bureau, in the first instance, must reconsider the rules we find infirm, Parts 11(B)— (C) infra, we perceive no useful purpose to be served by the Institute’s intervention at this point. We vacate the order in No. 80-1244, however, to eliminate any possible embarrassment to any effort by the Institute toward intervention in the future.
. See 5 U.S.C. § 706(2)(A) (1976).
. We find nothing in the Federal Alcohol Administration Act requiring a “hearing on the record,” see 5 U.S.C. §§ 553(c), 554(c)(2) (1976), which in turn would have called for a different standard of judicial review.
. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823-824 28 L.Ed.2d 136, 153 (1971).
. Almay, Inc. v. Califano, 187 U.S.App.D.C. 19, 26, 569 F.2d 674, 681 (1977).
. Home Box Office, Inc. v. FCC, 185 U.S.App. D.C. 142, 168, 567 F.2d 9, 35, cert. denied, 434 U.S. 829, 98 S.Ct. 111, 54 L.Ed.2d 89 (1977) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207, 216 (1962)). Accord, United States Lines, Inc. v. FMC, 189 U.S.App.D.C. 361, 368, 584 F.2d 519, 526 (1978).
. Ethyl Corp. v. EPA, 176 U.S.App.D.C. 373, 406-107 n.74, 541 F.2d 1, 34-35 n.74, cert. denied, 426 U.S. 941, 96 S.Ct. 2663, 49 L.Ed.2d 394 (1976).
. See text supra at notes 6-7.
. See text supra at notes 34-36.
. See, e.g., Atlantic Ref. Co. v. FTC, 381 U.S. 357, 367, 85 S.Ct. 1498, 1505, 14 L.Ed.2d 443, 451-452 (1965).
. See, e.g., Houston v. St. Louis Independent Packing Co., 249 U.S. 479, 484, 39 S.Ct. 332, 334, 63 L.Ed. 717, 720 (1919).
. The District Court held, and the consumers argue, that consumer-protection language in a labeling statute should be accorded its ordinary meaning. Wawszkiewicz v. Department of Treasury, supra note 5, 480 F.Supp. at 743 & n. 10, quoted in part in text supra at note 29; Brief for Consumers at 40-43. Certainly there are cases in which courts have read language in such legislation conformably with its common usage. See, e.g., 62 Cases of Jam v. United States, 340 U.S. 593, 599-600, 71 S.Ct. 515, 520, 95 L.Ed. 566, 571-572 (1951); United States v. 95 Barrels of Vinegar, 265 U.S. 438, 443, 44 S.Ct. 529, 531, 68 L.Ed. 1094, 1097 (1924); Armour & Co. v. Freeman, 113 U.S.App.D.C. 37, 39, 304 F.2d 404, 406, cert. denied, 370 U.S. 920, 82 S.Ct. 1559, 8 L.Ed.2d 500 (1962). In none of those cases, however, had an agency charged with enforcement of the statute offered a factually-supported rationale for a contrary interpretation. When agencies have done so, they have been upheld. See, e.g., Houston v. St. Louis Packing Co., supra note 49, 249 U.S. at 484, 39 S.Ct. at 334, 63 L.Ed. at 720; American Public Health Ass’n v. Butz, 167 U.S. App.D.C. 93, 97, 511 F.2d 331, 335 (1974) (term “false and misleading,” as used in 21 U.S.C. §§ 453(h)(12), 601(n)(12) (1976), “give[s] the Secretary discretion to determine what labeling, if any, will be required in addition to the official inspection stamp”). In Federation of Homemakers v. Butz, 151 U.S.App.D.C. 291, 466 F.2d 462 (1972), which involved a statute prohibiting “false or misleading” labeling, 21 U.S.C. § 607(d) (1976), we declined to uphold an administrative interpretation glossing statutory language, but only because “we [found] no basis in the record for the Secretary’s conclusion that [the language involved], in a label intended for the ordinary consumer, convey[s] the kind of technical and esoteric message that the Secretary finds in [it].” Id. at 465-466.
. See text supra at notes 32-33.
. 43 Fed.Reg. 37672 (1978). For the proposition that minimum percentage statements might mislead the consumer,' see, e.g., J.App. 375, 379, 383, 384, 386, 389, 392, 393, 396, 399-400, 407, 410, 420, 421-122, 437, 439, 445, 458. For the proposition that wine labels with minimum varietal percentages are not misleading, see, e.g., J.App'. 364, 365, 376, 381, 444.
. 43 Fed.Reg. 37672 (1978).
. 27 U.S.C. § 205(e)(1) (1976) (emphasis added), partially quoted in text supra note 2. See H.R.Rep.No. 1542, 74th Cong., 1st Sess. 12 (1935). , '
. 43 Fed.Reg. 37672 (1978).
. Id.
. See text supra at notes 8-9.
. 43 Fed.Reg. 37672 (1978).
. The Bureau’s explanation for its final rules on geographical labeling does not even mention the possibility of consumer deception. See id. at 37673-37674.
. 41 Fed.Reg. 50005 (1976). See text supra at note 17.
. See text supra at notes 10-11. Indeed, the Bureau withdrew, without explanation, its tentative plan to bar the use of geographically-suggestive corporate and trade names unless they had been approved prior to announcement of the proposed regulation that would have done so. See note 26 supra.
. 42 Fed.Reg. 30517 (1977).
. See, e.g., J.App. 83-85, 94.
. See text supra at note 12.
. See text supra at note 13.
. 5 U.S.C. § 553(c) (1976).
. 43 Fed.Reg. 37672, 37674 (1978).