Appellee brought this action in the District Court to compel disclosure under the Freedom of Information Act (FOIA), 5 U.S.C. 552, of certain information pertaining to eleven specifically identified research projects that had been approved and funded by the National Institute of Mental Health (NIMH), a unit of the Public Health Service of the Department of Health, Education, and Welfare (HEW). The eleven projects all involve research into the comparative effects of various psychotropic drugs on the behavior of children with certain learning disabilities. After in camera inspection of sample documents, the District Court ordered disclosure of all of the information sought, except that it contemplated that certain deletions might conceivably be made in respect of statements of opinion about the qualifications and competence of applicants for grants. The court further ordered the agency to amend its application instructions and regulations to conform with its decision. 366 F. Supp. 929. For the reasons hereinafter appearing, we affirm in part and reverse in part.

* The information in dispute is contained in three types of documents:

1. The Grant Application.

The initial grant application, among other things, identifies the research applicant, any research organization with which he may be affiliated, his qualifications and experience, the budget estimates, and the research protocol or design. Subsequent to the approval of the initial grant application, there may be filed continuation applications, renewal applications, and supplemental applications. Projects are approved for a specific 'project period' that may extend over several years, 42 C.F.R. 52.2(b), but a continuation application must be filed each year to report progress to date and justify support for the coming year. Id. 52.14(d). Renewal applications are required for periods beyond the originally scheduled project period, while supplemental applications are required for additional grants awarded because the amount previously awarded proves inadequate to carry out the project properly. Id.1

2. The Site Visit Report.

Outside consultants, engaged by HEW to review the grant application, frequently visit the location at which the research is proposed to be done, and thereafter prepare a report on their observations.

3. The Summary Statement ('pink sheet').

When the outside consultants have completed their work, an NIMH staff member assigned to them prepares a summary of their observations and deliberations and reports their recommendations. This statement will draw upon the site visit reports, if any.

The process by which applications are processed by NIMH and HEW, an understanding of which is necessary to appraise the significance of each type of document for FOIA purposes, is set out in considerable detail in the opinion of the District Court. Accordingly, we begin with only a brief recapitulation of how the process works.

Research of the type sponsored by NIMH is often of a highly sophisticated and specialized nature. In order to assure competent evaluation of each proposal, a system of so-called 'peer review' has been established, using the expertise of nongovernmental consultants functioning in panels organized around particular specialized disciplines within the broader field of biomedicine. These panels, called 'initial review groups' (IRGs), consist of from ten to twenty members, only one of whom, the Executive Secretary, is an NIMH employee.

Applications for NIMH research support are referred by the Executive Secretary to one member of the IRG as 'primary assignee,' and one or more other members with secondary responsibility. These assignees undertake to evaluate the application and gather such additional information as may be necessary to that task. This may involve a 'site visit' to the facility at which the applicant proposes to conduct his research. A site visit may be made, for example, in order to observe an experimental technique to be used in the proposed research.

An evaluation of each application, and a site visit report where applicable, are written by the assignee group and circulated to the whole IRG, together with the application, prior to its next meeting. (IRGs meet three times a year.) The application is discussed at length and a recommendation voted. If approval is recommended, it is also given a relative priority rating, since the cost of all proposals deemed worthy of funding may exceed the funds available.

Following the IRG meeting, the Executive Secretary prepares a Summary Statement for each application acted upon. The Summary Statement describes the proposal and recounts the substantive considerations that led the IRG to recommend approval or disapproval. It contains an opinion of the professional qualifications of the sponsor and an evaluation of his competence and facilities. The IRG's evaluation of the risk to human subjects, if any, is included, as is also a reference to the site visit report, if there is one. If there is a minority of two or more, the minority's view is also summarized, without attribution by name. The Executive Secretary may add a 'Note' in order to clarify any matter not resolved by the IRG, call attention to factors other than scientific merit, including policy considerations, or incorporate information obtained subsequent to the IRG meeting at which the application was considered.

Each application and the corresponding pink sheet is submitted to the National Advisory Mental Health Council (NAMHC),2 which was established to 'advise, consult with, and make recommendations to' the Secretary on Public Health Service activities in the field of mental health, 42 U.S.C. 218(c). The NAMHC is composed of three officials-- the Assistant Secretary for Health, the chief medical officer of the Veterans Administration, and a medical officer designated by the Secretary of Defense-- and twelve private citizens appointed by the Secretary on the basis of their qualifications in science, medicine, and/or public affairs. 42 U.S.C. 218(a).

The NAMHC may approve, disapprove, defer consideration of, or require additional IRG consideration of, any application. On occasion it does reject an IRG's recommendation of approval or disapproval,3 but ordinarily, instead of passing upon the scientific merits of each application, it gives primary attention to policy direction and emphasis, generally acting on applications in subject matter groups. Applicants are notified of the outcome, but only about 90% Of those receiving approval are actually funded by NIMH, to which the Secretary has delegated this function, due to limitations on funds. There is some ambiguity as to whether funding is determined solely on the basis of the ratings given by the IRGs and NAMHC, but the ratings may be presumed to be very influential in the funding decision.4

Each month NIMH makes public a list of all research grants awarded during the preceding month, including a general description of the project and its budget, and releases final progress reports received, except that release may be delayed up to six months pending publication by the researcher in a scholarly journal. The research design, proposed methods, and specific aims of a project are not made public, nor are the names or proposals of any applicants whose applications are disapproved.

The Freedom of Information Act requires discolsure, upon request, of the final opinions and identifiable records of each agency of the government, unless, in the case of the latter, they come within one of the nine specific exemptions in the Act. The burden of proof is on the agency opposing disclosure, and the exemptions therefrom are to be narrowly construed. The government relies upon three sparate exemptions to justify nondisclosure of the various types of information here sought, as follows:

1. Exemption 4, for trade secrets and commercial or financial information received in confidence, is invoked to cover the research designs submitted in applications and described in the Summary Statements, site visit reports, and progress reports.

2. Exemption 5, for inter-agency or intra-agency memoranda that would not be subject to discovery in litigation, is said to cover the Summary Statements and site visit reports in their entirety, except insofar as purely factual information is involved.5

3. Exemption 6, which applies to personnel, medical, and 'similar' files the disclosure of which would be a clearly unwarranted invasion of personal privacy, is raised with respect to statements of opinion in the Summary Statements and site visit reports as to the professional qualifications and competence of applicants who received grants.

A. Exemption 4.

The essence of the argument that the research designs submitted in the expectation of confidentiality are trade secrets or commercial information is that 'ideas are a researcher's 'stock-in-trade." Their misappropriation, which, it is claimed, would be facilitated by premature disclosure, deprives him of the career advancement and attendant material rewards in which the academic and scientific market deals, in much the same way that misappropriation of trade information in the commercial world deprives one of a competitive advantage. Indeed, the government has been at some pains to argue that biomedical researchers are really a mean-spirited lot who pursue self-interest as ruthlessly as the Barbary pirates did in their own chosen field. Whether this is the sad truth, or whether, as appellee suggests, 'secrecy is antithetical to the philosophical values of science,' is not, however, an issue in this case; the reach of the exemption for 'trade secrets or commercial or financial information' is not necessarily coextensive with the existence of competition in any form.

It is clear enough that a noncommercial scientist's research design is not literally a trade secret or item of commercial information, for it defies common sense to pretend that the scientist is engaged in trade or commerce.6 This is not to say that the scientist may not have a preference for or an interest in nondisclosure of his research design, but only that it is not a trade or commercial interest. To the extent that his interest is founded on professional recognition and reward, it is surely more the interest of an employee than of an enterprise,7 and we are far from persuaded that Congress intended in Exemption 4 to apply terms drawn from the business context to the employment market.8 We cannot, consistently with the Act's recognized mandate to construe exemptions narrowly, see Vaughn v. Rosen, 157 U.S.App.D.C. 340, 484 F.2d 820, 823 (1973), cert. denied 415 U.S. 977, 94 S. Ct. 1564, 39 L. Ed. 2d 873 (1974); Getman v. NLRB,146 U.S.App.D.C. 209, 450 F.2d 670, 672, stay denied, 404 U.S. 1204, 92 S. Ct. 7, 30 L. Ed. 2d 8 (1971), extend them by analogies that lead so far away from the plain meaning of Exemption 4. Consequently, we hold that research designs submitted in grant applications are not exempt from disclosure under the Act. This holding extends to all types of applications-- initial, continuation, supplemental, and renewal-- and to progress reports made by grantees as part of the last three kinds of applications.

B. Exemption 5.

The applicability of Exemption 5 to the site visit reports made by members of the IRG, and to the Summary Statement written by the Executive Secretary to report on the IRG's recommendation to the NAMHC, turns on whether the IRG is an 'agency' under the Administrative Procedure Act, of which the FOIA is a part. If the IRG is indeed an agency, then appellee's position that the Summary Statements and accompanying site visit reports constitute its final opinions, which must be made available, 5 U.S.C. 552(a)(2)(A), is not without force. If, on the other hand, the IRG is not an agency but merely a unit within another agency, then these documents are identifiable records in the hands of that agency and it is our task to determine whether they are exempt from disclosure as intra-agency memoranda.

1. The IRG as an agency vel non.

The APA defines the term 'agency' to mean 'each authority of the Government of the United States, whether or not it is within or subject to review by another agency,' with certain specific exceptions not applicable in this case. 5 U.S.C. 551(1). The generality of this definition has required the commentators that have dealt with it to attempt an elaboration along more functional lines than the phrase 'each authority' conveys,9 but recent cases have made it clear that any general definition can be of only limited utility to a court confronted with one of the myriad organizational arrangements for getting the business of the government done. See Grumman Aircraft Engineering Corp. v. Renegotiation Board, 157 U.S.App.D.C. 121, 482 F.2d 710 (1973) (Regional Boards) (hereinafter Grumman II), cert. granted, 417 U.S. 907, 94 S. Ct. 2602, 41 L. Ed. 2d 211 (1974); Soucie v. David, 145 U.S.App.D.C. 144, 448 F.2d 1067 (1970) (Office of Science and Technology); International Paper Co. v. FPC, 438 F.2d 1349 (2d Cir. 1971) (Staff of FPC); Larche v. Hannah, 177 F. Supp. 816 (W.D.La.1959) (Civil Rights Commission), rev'd on other grounds, 363 U.S. 420, 80 S. Ct. 1502, 4 L. Ed. 2d 1307 (1960). The unavoidable fact is that each new arrangement must be examined anew and in its own context.

Congress has authorized the Secretary of HEW to make only such mental health and medical research grants 'as are recommended' by the various National Advisory Councils, in this case the NAMHC, 42 U.S.C. 241(d), (i), that it established. 42 U.S.C. 218.10 It also, however, authorized the Secretary to 'appoint such advisory committees (in addition to those authorized to be established under other provisions of law) . . . as he deems desirable for the purpose of advising him in connection with any of his functions,' 42 U.S.C. 217a(a), and to compensate nongovernmental members thereof. Id. at 217a(b).11

The authority to establish advisory committees by administrative action was first exercised in the period after World War II, when the vast expansion of supported biomedical research made it 'impractical to encompass in a council membership the expertise in all the numerous disciplines, fields, specialties, and areas represented in biomedical research proposals . . .. To overcome these deficiencies, initial review groups were established to assist the councils.' S.Rep. No. 381, 93d Cong., 1st Sess. 38 (1973). In 1968, when NIMH was established as a distinct part of HEW, this system was carried over with the creation of the NAMHC, and the Institute's decision to create the IRGs. Hearings, supra note 4, at 3620-21. The NIMH dual review process remains, however, 'less formal' than that of the NIH: 'IRGs are established or phased out as required by the size and nature of the grant review workload . . ..' S.Rep. No. 381, supra, at 38. Ad hoc IRGs are formed to advise on particular applications 'not within the competence' of any then-standing group. NIHM Handbook, supra note 2, at 9.

Under the prior decisions of this court, we think the IRGs are advisory committees, performing staff functions through the medium of outside consultancy, and are not agencies. The considerations raised in Soucie v. David, supra, and Grumman II, supra, point unmistakeably to this conclusion, which comports with our present impression of how this question should be handled Soucie involved the status of the Office of Science and Technology, which, in addition to advising and assisting the President in coordinating federal policy for science and technology, was also authorized, as we emphasized, independently to evaluate federal programs.12 It was created by an executive reorganization plan submitted to Congress and 'explicitly considered' by the House. Id. at 1074. Congress approved the plan with the understanding that it was 'delegating (to OST) some of its own broad power of inquiry in order to improve the information of federal scientific programs available to the legislature.' Id. at 1075. It 'clearly contemplated,' as did the Executive, 'that the OST would function as a distinct entity and not merely as part of the President's staff.'13 Finally, we noted that since the OST had published FOIA regulations, it had apparently considered itself, prior to the litigation, to be an agency subject to the APA-- a consideration of some weight. Udall v. Tallman, 380 U.S. 1, 16, 85 S. Ct. 792, 13 L. Ed. 2d 616 (1965). All of these factors taken together led to the determination that OST exercised 'substantial independent authority,' and to the conclusion that it was an agency subject to the Freedom of Information Act.

Some of the same factors considered in Soucie were present in Grumman II, which dealt with the status of the Regional Boards established by the Renegotiation Board. The Regional Boards' exercise of formal decision making power delegated from the National Board was found to be 'within Congress' contemplation when it established the Renegotiation Board apparatus.' Indeed, the statutory authorization there at issue was expressly that of delegating functions 'to any agency of the Government, including any such agency established by the Board.' 50 U.S.C.App. 1217(d). Further, the National Board had promulgated FOIA regulations for disclosure of some Regional Board documents, again indicating its own view, prior to litigation, that they were agencies subject to the FOIA. Furthermore, the Regional Boards fell squarely within the analytic definition supplied by Professor Freedman, whose consideration of this question is the fullest and most discerning, namely, they were the body in which 'substantial 'powers to act' with respect to individuals (were) vested.' Freedman, supra note 9, at 9. They had their own staffs to investigate and negotiate with contractors, and their 'recommendation is communicated openly to the contractor prior to any assumption of jurisdiction by the National Board.' 482 F.2d at 715. They were, in short, the agency with which an affected member of the public dealt, and from whose decision appeal might lie, depending upon the amount in controversy.

The contrast between the agencies involved in Soucie and Grumman II, on the one hand, and the IRGs, on the other, could not be greater. Unlike the OST, the IRGs do confine themselves to making recommendations; authority to make grants is vested in the Secretary, and authority to recommend doing so lies with the NAMHC. The IRGs act as consultants to the NAMHC; their members are strictly forbidden from communicating their group's recommendations to applicants. NIMH Handbook, supra note 2, at 30. Applicable regulations and administrative rules within NIMH and HEW have consistently reflected the view that the IRGs are not subject to the FOIA, see id.; 45 C.F.R. 5.72(b), 5.73(a); and the authority under which they are appointed gives no hint of a congressional expectation that NAMHC's delegation of the initial review function would somehow make either more or less information available to the legislature or the public.

Clearly, the work now done by IRGs could again be done by the NAMHC if it sat continuously instead of meeting three times a year for about two days each time. Employing consultants to improve the quality of the work that is done cannot elevate the consultants to the status of the agency for which they work unless they become the functional equivalent of the agency, making its decisions for it. There is no doubt in this instance that the usually perfunctory review the NAMHC gives to a particular application-- as opposed to the group of which it is a part-- makes the IRG's recommendations an often crucial element in the approval process. But, just as the APA makes the fact that a government authority's decisions are subject to review irrelevant in determining whether that authority is an agency, at least in this case the degree of scrutiny its decisions are given on review is equally beside the point.

The important consideration is whether it has any authority in law to make decisions. The IRGs have not; their favor is not necessary because the law empowers the Secretary to make grant awards if (and only if) the NAMHC so recommends. The fact that the NAHMC may be greatly influenced by the IRG's expert view does not make the IRG an agency. See International Paper Co. v. FPC, 438 F.2d 1349, 1359 (2d Cir. 1971). In Soucie this court did not consider whether the President generally accepted the advice of OST. In that case and in Grumman II we looked to the functions that OST and the Regional Boards respectively were empowered by law to perform. The alternative would inevitably involve the courts in determining the care with which the decisional officers of government agencies supervise their staffs, in order to determine whether in fact the staff is not the stand-in for the officer or agency-- its recommendations his decisions-- and whether it ought not therefore in law to be held accountable as the agency. This we cannot do. Morgan v. United States, 304 U.S. 1, 18, 58 S. Ct. 999, 82 L. Ed. 1129 (1938); United States v. Morgan, 313 U.S. 409, 422, 61 S. Ct. 999, 85 L. Ed. 1429 (1941).

That is not to say that a staff recommendation may never achieve the dignity of an agency's final decision; it may do so when the agency adopts it as its own, and at that point its disclosure can be required. American Mail Line, Ltd. v. Gulick, 133 U.S.App.D.C. 382, 411 F.2d 696 (1969), is a case in point. There the question was 'whether an administrative agency (the Maritime Subsidy Board) may take affirmative action against a private party by means of a decision in which it states that the only basis for such action is a certain specified (staff) memorandum and then refuse to disclose the memorandum to the party affected by the action.' While the question was answered in the negative, the reason was that the agency had made the memorandum the express and only basis for its decision and not that it had made the staff into the 'agency.' In the present case, however, leaving aside the problem of whether even the denial of a grant award is 'affirmative action' against an applicant, the reasoning of the IRG is not the 'only basis' for the Secretary's decision. The Executive Secretary's Note and the NAMHC's policy choices, and in some instances the latter's particularized scrutiny, intervene.14 Cf. Sterling Drug, Inc. v. FTC, 146 U.S.App.D.C. 237, 450 F.2d 698, 706 (1971). We hold, therefore, that the IRG is not itself an agency under the APA nor, consequently, subject to the strictures of the FOIA.15

2. Application of Exemption 5.

Having decided that the IRG is not an agency, nor its Summary Statements and site visit reports the final decisions of an agency, it becomes necessary to determine just how much of the disputed information in these intraagency documents is exempt from disclosure. Exemption 5 applies only to matters 'which would not be available by law to a party other than an agency in litigation with the agency.' While there are often problems in determining the precise scope of the exemption without the benefit of actual litigation the nature of which informs the breadth of discovery, see Environmental Protection Agency v. Mink, 410 U.S. 73, 86, 93 S. Ct. 827, 35 L. Ed. 2d 119 (1973), its application in this case is relatively uncomplicated. As a general proposition Exemption 5 does not shield from disclosure 'purely factual, investigative matters,' as opposed to 'materials reflecting deliberative or policy-making processes.' Id. at 89, 93 S.Ct. at 837. Even purely factual matter may be exempt, however, if it is inextricable without compromise of the deliberative process, id. at 91, 93 S. Ct. 827, and so too may be 'a summary of factual material (that) is part of the deliberative process,' even though the facts themselves are elsewhere on the public record. Montrose Chemical Corp. v. Train, 160 U.S.App.D.C. 270, 491 F.2d 63 (1974).

In order to apply these propositions to the facts at bar, the contents of the Summary Statements and site visit reports need further elaboration at this point. We take as our texts appellee's Exhibit 1, the NIMH Handbook, supra note 2, at 33-36, documents as illustrated by the sample submitted to the court. The Summary Statements begin with (1) a concise resume, 'no more than six or seven sentences,' of the proposed project, its review by the IRG, and the reasons for the IRG's recommendation, including the contrary reasons offered by a minority.16 There follows (2) a 'brief description of the proposal,' its 'aims, methodology, and, for renewal, supplemental and revised proposals, the background or history.' The next and 'most critical' section is the IRG's critique (3) which discusses 'the strengths and weaknesses of various aspects of the proposal in detail.'17 The 'background and competence' of the applicant and his associates (4) are discussed, as are (5) 'any special aspects of the facilities and equipment and the extent of departmental and interdepartmental cooperation' at the applicant's institution. The proposed project budget is then analyzed with reference to its adequacy, justification, and projected duration (6). Supplemental requests are related to previously approved amounts.18 To all of this may be appended (7) the Executive Secretary's Note, described supra at p. 242, (8) the minority report of two or more dissenting members, and (9) a summary of any site visit report.

The site visit report itself contains, in addition to purely identifying material, such as application number, date, and persons seen, (10) evaluations of the proposal, the investigator, and his staff, (11) sections on the facilities and other support available at the institution and (12) the budget, and (13) 'other comments.' Because of the substantial overlap necessary between the site visit report and Summary Statements, site visitors are advised to follow the format for the latter document 'since the site visit report, if accepted by the (IRG), can serve as a basis for the Summary Statement.' NIMH Handbook, at 46.

From this mere recitation it is clear that most of the matters called for in the site visit report and Summary Statement for each application are evaluative, and call into play the policy of protecting the deliberative process, at which Exemption 5 is directed. See EPA v. Mink, supra; Soucie v. David, supra. Indeed, the only matters that are even arguably subject to compelled disclosure are the Summary Statements' (2) description of the proposal, its aims and methodology, and any factual matter contained in (9) the summary of the site visit report. In the site visit report itself, only (11) the statement of facilities, and (12) the budget, merit comment.

Of these four items, the two (2 and 9) in the Summary Statement are abstracts of other information-- either the site visit report or portions of the underlying application. As such we think them covered by the reasoning of Montrose Chemical, supra. That case involved application of Exemption 5 to summaries, made by agency staff attorneys, of evidence developed at a public hearing. The summaries were prepared for and submitted to the Administrator of the Environmental Protection Agency 'to assist (him) in his study of the record' on the basis of which he was obliged to make a decision. This court held the summaries exempt as an integral part of the deliberative process. Sensitive to the necessity of attaching varying degrees of significance to different facts in the course of epitomizing the record, we said:

Even if they cited portions of the evidence verbatim, the assistants were making an evaluation of the relative significance of the facts recited in the record; separating the pertinent from the impertinent is a judgmental process, sometimes of the highest order; no one can make a selection of evidence without exercising some kind of judgment, unless he is simply making a random selection. 491 F.2d at 68.

No significant difference distinguishes the present case from Montrose Chemical. The research design and description of methodology in the application submitted for in camera inspection covers fifteen, single-spaced typewritten pages; their description in the Summary Statement is one page in length.19 In the Montrose Chemical paradigm, the judgmental element arises through the necessity to select and emphasize certain facts at the expense of others. In the instant case, where the whole proposal must be described at least in general, various aspects of its are described in greater detail than others. In virtually every sentence the author must operate at a level of specificity that reflects his personal perspective on the material being summarized.

An example may be taken from the Summary Statement before us without compromising any information for which exemption is claimed. In the course of describing a proposal for evaluating the relative efficacy of a number of drugs in the treatment of hyperkinetic children, the following statement appears: 'The assessment battery consists of a number of rating scales, and various cognitive and performance measures.' Significantly, however, while the various performance and cognitive tests are then enumerated, the rating scales are neither identified nor described, but are said only to have proved sensitive in prior, unspecified studies. This difference of treatment may well reflect no more than the greater ease with which named tests can be referenced than rating scales described. It may, on the other hand, reflect the view that certain well-known tests have a definite reputation for reliability, and that rating scales are only so much surplusage. A different group of reviewers with a different set of views might well have elaborated on the rating scales' utility and never specified the cognitive and performances tests proposed. What the effect of such a choice might be on the proposal's prospects we need not guess. The point is simply that choices are and must be made by someone or some group with a unique perspective, and decisions may be based on them. Accordingly, the two items under discussion must also be held exempt from disclosure.20

The NIMH Handbook alone does not indicate whether the two items in the site visit report-- (11) the facilities and (12) budget references-- are meant to be narrative or analytic. The representative site visit report submitted to the District Court, which is of controlling significance,21 suggests the latter, however, at least with respect to budgetary considerations. The short section on the budget relates the amount requested to the site visitors' analysis of the amount needed, and suggests for IRG consideration a possible economizing step. As such, it is clearly a part of the deliberative process and exempt from compelled disclosure. But this particular report makes no reference to the matter of facilities, neither describing nor rendering an opinion on the adequacy of the facilities available to the grant applicant.22 The entire document, however, other than the opening paragraphs which describe the proposal and for which no claim of exemption is made, is an expression of the visitors' opinions and not a recitation of facts. This suggests that facilities references, if any, in the site visit reports for other of the applications sought by appellee would be of the same nature. In light of the parties' agreement, supra note 21, however, we need not choose between reliance on this speculation, with its potential effect of relieving the agency of its statutory burden of proof, and in camera inspection of all the site visit reports in suit. Finding no matter of the type sought in the controlling document, there is no relief respecting item (12) to which appellee can lay claim.23

Appellant challenges the District Court's jurisdiction to order the agency to amend its regulations to conform with the court's opinion. The FOIA, it is contended, 'grants jurisdiction to the district courts only to review agency denials of requests for specific documents and to enjoin withholding of those documents' from the person who made the request. That is of course true insofar as it goes, but is not responsive to whether the court may not draw on powers apart from, and unabridged by, the FOIA in order to give complete relief where it is due. 'With the express vesting of equitable jurisdiction in the District Court by 552(a), there is little to suggest, despite the Act's primary purpose, that Congress sought to limit the inherent powers of an equity court.' Renegotiation Board v. Bennercraft Clothing Co., 415 U.S. 1, 94 S. Ct. 1028, 39 L. Ed. 2d 123 (1974) (dictum).

One can imagine circumstances, such as where an agency simply refuses to conform its actions to the known requirements of the Act in order to deter requests for information by repetitive litigation, that would tempt a court to use any or all of 'the usual weapons in the arsenal of equity,' Bannercraft Clothing Co. v. Renegotiation Board, 466 F.2d 345, 354 (1972), rev'd on other grounds, 415 U.S. 1, 94 S. Ct. 1028, 39 L. Ed. 2d 123 (1974). In the case at bar, however, it is unnecessary to decide whether the District Court would be so empowered.

Appellee initiated the process culminating in this action by a letter requesting access to documents relating to eleven specifically identified research grants. When the request had been denied in part and administrative appeal exhausted, appellee filed a complaint the prayer of which requested that the court declare the plaintiff's right to disclosure of the disputed records and order their disclosure, and 'that this Court declare invalid under the Freedom of Information Act the regulations issued by (HEW) which exempt from public disclosure all research protocols and all proposed grant applications.' JA6. In its opinion the District Court merely suggested that 'at a minimum, the defendants should promptly modify existing regulations and grant application instructions to bring them into conformity with the decision of this Court,' but its order elevated this suggestion into an injunctive obligation presumably enforceable in the same manner as any injunction, namely, by contempt.

The FOIA requires each agency to make information, not exempt by the terms of the Act, available 'in accordance with published rules.' From this may readily be inferred an obligation to publish rules that accurately reflect the agency's substantive obligations under the Act, and rules that fail to do so are of no force when 'any person' seeks access to information not exempt from disclosure under the Act. Pretermitting the very real question of whether a single request for documents creates a continuing case or controversy24 sufficient to support an order to amend regulations of only speculative future effect on an FOIA plaintiff, there is no warrant in the record of this case for anticipating that HEW would not proceed in good faith to incorporate the substance of a final court decision into its rules and practices.

The District Court was sensitive to the public interest that the FOIA, 'to the extent practical, be self-operative to insure prompt disclosure.' It was equally aware of the necessity 'that grant applicants be placed on notice that information submitted pursuant to an application for NIMH grant funds' is subject to public disclosure. We, of course, share the court' concern, but are without sufficient reason to doubt that appellant does also. Considerations of inter-branch comity impel us to withhold coercive orders that are not demonstrably necessary. Cf. Nixon v. Sirica, 159 U.S.App.D.C. 58, 487 F.2d 700, 712 (1973).

What we have held hereinabove is that the eleven initial grant applications involved in this case (all of which had been approved by HEW), together with any continuation, renewal, or supplemental applications incident thereto (either approved or pending), are not exempt from disclosure under the Freedom of Information Act. Contrarily, we have held that site visit reports and Summary Statements are exempt under Exemption 5. The impact of this latter holding is limited in this case by the fact that HEW has voluntarily disclosed the purely factual matter contained therein, in an apparent recognition that such matters do not come within the purposes of the exemption. Lastly, we have found, in the circumstances of this record, an inappropriate exercise of equity jurisdiction in the District Court's injunctive command that HEW conform its regulations to the court's mandate.

The judgment of the District Court is, accordingly, affirmed in part and reversed in part; and the case is remanded for the entry of a decree consistent herewith.

It is so ordered.