McGOWAN, Circuit Judge:

This appeal by the Government from a declaration and order of the District Court presents the question of the extent to which press communication with federal prisoners can be restricted without falling afoul of the First Amendment. It arises by reason of the issuance and implementation by the Federal Bureau of Prisons of its Policy Statement 1220.1A, which categorically forbids personal interviews by newsmen of all prisoners in all of the correctional institutions administered by the Bureau. Because of the great respect which the federal judiciary entertains for the Bureau by reason of its long and continuous history of distinguished and enlightened leadership, both the District Court and ourselves have been at special pains to assure the development of an eviden-tiary record adequately illuminative of the important issue to be resolved. By reference to that record, and the supportable findings and conclusions carefully distilled therefrom by the District Court, we affirm the action taken by it. That action is essentially to require that the Bureau reformulate its policy in less absolute terms, thereby reducing the unacceptable degree of tension which now exists between the policy and the First Amendment, and striking a justifiable balance between the administrative needs of the Bureau and the constitutional guarantee of the public’s right to know about the public business through the functioning of a free press.

Appellees are a major newspaper and one of its reporters. In March of 1972 the reporter requested of the Director of the Bureau permission to conduct interviews in the federal prisons at Lewis-burg, Pennsylvania and Danbury, Connecticut. The inmates he sought to interview were members of inmate negotiating committees that had been formed during work stoppages during the prior month, as well as certain other inmates who had written him to complain of their treatment. Relying on Policy Statement 1220.1A, the Director rejected the requests. 1

Suit was then filed in the District Court challenging these denials and the Policy Statement on which they were based. The District Court declared that the Bureau’s Policy Statement, insofar as it flatly prohibited all press interviews, violated the First Amendment; and it ordered the Bureau to stop enforcing that policy, and, pending its modification, to proscribe only those requested press interviews that are likely immediately and directly to cause serious administrative or disciplinary problems. Washington Post Co. v. Klein-dienst, 357 F.Supp. 770, 779 (D.D.C. 1972). 2

*997 The Supreme Court stayed the District Court’s order pending resolution of the appeal to this court. 406 U.S. 912, 92 S.Ct. 1761, 32 L.Ed.2d 112 (1972). After oral argument and while the appeal was under submission, we remanded the case to the District Court, suggesting the desirability of expanding ■ the record by further evidentiary hearings in order that findings of fact could be made on specified issues, and the conclusions of law reexamined in light of the more developed factual record and the intervening Supreme Court discussion of press freedoms in Branzburg v. Hayes, 408 U.S. 665, 92 S.Ct. 2646, 33 L.Ed.2d 626 (1972), which had been pressed upon us by the Government. 155 U.S. App.D.C. 283, 477 F.2d 1168 (D.C.Cir. 1972). On remand, the District Court held further evidentiary hearings, enlarged its findings of fact, and examined Branzburg and other recent decisions urged upon it by the parties. The District Court reaffirmed its initial decision, Washington Post II, 357 F.Supp. 779 (D.D.C.1972), and the case returned to this court where it was supplementally briefed and orally argued a second time.

The Bureau’s policy governing press interviews differs significantly from that controlling the visitation rights of other persons. In general, inmates’ families, their attorneys, and religious counsel are accorded liberal visitation privileges. Even friends of inmates are allowed to visit, although their privileges appear to be somewhat more limited. The testimony suggests that the federal institutions follow a rather liberal policy of granting visitation privileges whenever possible. 3

Policy Statement 1220.1A, promulgated on February 11, 1972, establishes the Bureau policy controlling all means of news gathering within the federal institutions. Under that policy, inmates are permitted to correspond freely with members of the news media. The correspondence is funneled through sealed prisoners’ mailboxes provided for the purpose of assuring that no censorship occurs. Likewise, representatives of the news media are permitted to initiate or follow up correspondence with inmates of their choosing, and incoming letters are only inspected to assure that they do not contain contraband or materials that would incite illegal conduct. 4 Additionally, the Bureau’s proclaimed policy is to encourage visitation by representatives of the news media, by advance appointment, for the purpose of preparing reports concerning institutional facilities, programs, and activities. 5

The Bureau’s policy governing private press interviews — the focus of this litigation — is set forth in the first two sentences of paragraph 4(b) (6) of the Policy Statement:

Press representatives will not be permitted to interview individual inmates. This rule shall apply even where the inmate requests or seeks an interview. However, a conversation may be permitted with inmates whose identity is not to be made public, if it *998 is limited to the discussion of institutional facilities, programs and activities.

The Bureau’s distinction between an “interview” and a “conversation” was explored in the evidentiary hearings and reflected in the District Court’s findings of fact. Essentially, the Bureau considers a “conversation” to be a spontaneous discussion with inmates whom a newsman might randomly encounter during a supervised inspection tour of the institution. Such conversations are limited in time to approximately five to ten minutes, and are restricted by the terms of the Policy Statement to the discussion of “institutional facilities, programs and activities.” Additionally, reporters who rely on information received in a “conversation” are requested not to name the inmates with whom they spoke. By contrast, the Bureau considers an “interview” to be a private, scheduled, face-to-face discussion with a designated inmate that lasts a sufficient time to permit extensive discourse. 6

The Policy Statement’s prohibition against interviews is total. They are to be denied regardless of the characteristics and record of the inmate; his desire to be interviewed; and the conditions prevailing at the institution at the time the interview is sought. Moreover, the interview ban applies with equal force to all federal institutions, minimum as well as maximum security, and even precludes " interviews with offenders who are employed in business establishments or attending schools pursuant to community release programs.' 7 '

Appellees assert a First Amendment right of their own to gather and report news, together with a parallel First Amendment right of the public to have them serve that function, as the basis for relief from the strictures of the Policy Statement. 8 While acknowledging that the Bureau has provided some scope for gathering news in the federal institutions, appellees maintain that it is inadequate, and that person-to-person interviews are essential to the effective performance of their journalistic function in the prison setting.

The First Amendment rights of the press are, of course, a function of the paramount right of the public to information necessary to enable it to assert ultimate control over the political process. 9 Although this is by now axiomatic, it is particularly important to be borne in mind in considering the regulation of press access to information concerning the administration of the correctional system. Courts must be alert to assure that the diminution of informa *999 tion in this area — one of increasing public concern — is no greater than is necessary for the protection of the legitimate societal interests in the effective administration of those systems.

Right of access by the press to newsworthy events is necessarily antecedent to its First Amendment right to publish. Obviously, excessive limitations on the ability to gather information would render the freedom to publish a hollow guarantee. This concept is not novel. The Supreme Court’s observation in Ze-mel v. Rusk, 381 U.S. 1, 17, 85 S.Ct. 1271, 1281, 14 L.Ed.2d 179 (1965), that the “right to speak and publish does not carry with it the unrestrained right to gather information,” acknowledges, as Justice Stewart has since noted, that some right of access does exist. Branz-burg v. Hayes, supra, at 728, n. 4 of 408 U.S., 92 S.Ct. 2646 (Stewart, J., dissenting). Indeed, one point on which all of the Justices agreed in Branzburg was that the right of the press to gather information is one that must be afforded some constitutional recognition. As Justice Stewart observed:

A corollary to the right to publish must be the right to gather news. The full flow of information to the public protected by the free-press guarantee would be severely curtailed if no protection whatever were afforded to the process by which news is assembled and disseminated.

Id. at 727, 92 S.Ct. at 2672. See also, id. at 715, 92 S.Ct. 2646 (Douglas, J., dissenting); Note, The Right of the Press to Gather Information, 71 Colum. L.Rev. 838 (1971).

Like the right of the public to know, the right of the press to access to newsworthy events is not without limits. Zemel, supra, at 17 of 381 U.S., 85 S.Ct. 1271. News gatherers are excluded from a number of governmental processes, such as grand jury proceedings, the deliberative conferences of judges, and closed executive and congressional inquires. Nor do newsmen have a constitutional right of access to the scene of a crime or disaster when the general public is excluded, for, as Justice White has pointed out, “the First Amendment does not guarantee the press a constitutional right of special access to information not available to the public generally.” Branzburg, supra, at 684 of 408 U.S., at 2658 of 92 S.Ct. However, both logic and Supreme Court expressions suggest that the converse is equally true, that is to say, the right of access of the press to report an event is customarily as broad as the general public’s access to witness it. See Estes v. Texas, 381 U.S. 532, 540, 85 S.Ct. 1628, 14 L. Ed.2d 543 (1965). Restrictions that single out members of the news media and limit their access more grudgingly than the right granted the public at large should be examined with care and upheld only where the differential treatment is in response to an identified evil associated with press attendance. 10

The inquiry in the present case does not end with the Government’s assertion of the truism that prisons are institutions where public access is generally limited. Appellees do not seek carte blanche rights freely to enter federal institutions whenever they wish and to converse with any person of their choosing. Rather, they say that they should be afforded essentially the same limited rights granted the public generally — the right on occasion to enter and engage in private interviews with consenting inmates.

The Government also maintains that, in light of the alternative means of access provided the press by the Policy Statement, the prohibition against press interviews does not constitute any abridgment of the First Amendment rights of the press; since the flow of information is sufficiently assured by *1000 press inspections and incidental conversations, and by the virtually unlimited ability to communicate by mail, the Bureau can protect its legitimate interests by imposing a total ban on press interviews. However, as the Supreme Court said in Kleindienst v. Mandel, "[t]his argument overlooks what may be particular qualities inherent in sustained, face-to-face debate, discussion and questioning.” 408 U.S. 753, 765, 92 S.Ct. 2576, 2583, 33 L.Ed.2d 683 (1972). More importantly, the Government’s underlying factual premise was rejected in the District Court’s findings of fact.

One of the specific questions posed by us on remand was the extent to which the accurate and effective reporting of news is critically dependent on the opportunity for private personal interviews. In the hearing following remand, the original testimony of appellee Bagdikian was supplemented at length by testimony of other persons having journalistic experience. 11 Mr. Arthur L. Liman, General Counsel of the New York State Special Commission on Attica, who personally supervised perhaps the most extensive series of interviews ever conducted with prison inmates, also testified. On the basis of this substantial and uncontroverted evidence, the District Court found that the sources of information provided by Policy Statement 1220.1A were inadequate to permit the news media to develop an accurate and penetrating knowledge of prison conditions or events, and that accurate and effective news reporting about prison conditions is critically dependent on the opportunity for personal interviews with the inmate population. Washington Post II, 357 F.Supp. at 781.

The evidence disclosed a number of characteristics common to news reporting in general, as well as certain factors unique to the prison setting, that make personal interviews particularly critical to accurate reporting in that context. In general, the testimony indicated that responsible journalists are reluctant to publish a story without first speaking personally to the source of the information in order to assess his credibility and reliability. Even when the story will not be attributed to the person from whom it came, journalists consider this form of verification to be an essential element of responsible reporting.

The general importance of personal inquiry in order to check a news story is heightened in the context of prison institutions. The,percentage of functional illiterates in the inmate population tends to be high. This minimizes the effectiveness of the mail services as a vehicle for news gathering. Many of the inmates simply cannot communicate by that medium; some of the others who try are only able to read press inquiries and compose responses with the assistance of the more educated inmates. The literacy problem aside, communication by mail lacks the spontaneity and flexibility of a personal interview, and denies the reporter the ability to follow a thought. A personal interview gives feel and depth perspective.

Nor does the Bureau’s policy of permitting press tours and conversations incidental thereto appear to plug the information gaps produced by the prohibition of press interviews. Conversations initiated pursuant to press tours are of necessity random and sporadic in nature, and are considerably more limited in *1001 time and permissible scope of discussion. 12 Moreover, the tour conversations often occur with groups rather than with individual inmates. This both inhibits the newsman’s ability to make a careful assessment of the credibility of individual inmates and imposes significant peer group pressures on inmates to inflate or adapt their stories to correspond to the perceived desires of their fellow prisoners. 13

The District Court’s findings on the critical importance of interviews to effective news reporting of prison matters are supported by a substantial body of evidence of record, and indeed appear to be uncontradicted. We agree with the District Court’s conclusion that personal interviews are essential to effective news reporting in the prison environment. In this context, we are mindful of Justice Blackmun’s admonition in Kleindienst v. Mandel, supra, at 765 of 408 U.S., 92 S.Ct. 2576, that the existence of alternative means of communication provided the press by the Bureau does not necessarily dispose of the question before us. Alternative means of communication are a relevant, but not conclusive, factor in our consideration.

Ill

We emphasize that we are not confronted here with the claim that the Bureau can under no circumstances prohibit press interviews. Clearly the requirements of the Bureau often can, in an individual context, justify substantial curtailment of First Amendment protections. The issue before this court is rather one of overbreadth — whether Policy Statement 1220.1A exacts a greater toll than is justified by the admittedly legitimate interests the Bureau seeks to protect.

The Supreme Court has repeatedly cautioned that regulations touching on First Amendment freedoms must be narrowly drawn, reflecting a “principle that justifiable government goals may not be achieved by unduly broad means having an unnecessary impact on protected rights of speech, press, or association.” Branzburg, supra, at 680-681 of 408 U.S., at 2656 of 92 S.Ct. As the Court pointed out in N. A. A. C. P. v. Button, “Broad prophylactic rules in the area of free expression are suspect^ Precision of regulation must be the touchstone in an area so closely touching our most precious freedoms.” 371 U.S. 415, 438, 83 S.Ct. 328, 340, 9 L.Ed.2d 405 (1963) (Citations omitted). See also Kusper v. Pontikes, 414 U.S. 51, 94 S.Ct. 303, 38 L.Ed.2d 260 (1973); Grayned v. City of Rockford, 408 U.S. 104, 119, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972); Women Strike for Peace v. Morton, 153 U.S.App.D.C. 198, 472 F.2d 1273 (1972).

Examined in this context, the question becomes whether the interests of the Bureau justify the absolute ban on all press interviews within all of the diverse correctional institutions of the federal system. Recognizing the difficulties inherent in placing quantitative values on the interests served by the Bureau’s broad ban on interviews, and the special sensitivity involved in examining the Bureau’s judgments in prison administration, we nonetheless are bound to ask whether the Bureau has advanced “any substantial regulatory interest which can justify the broad *1002 prohibitions which it has imposed.” 14 We turn to the justifications offered to support the Bureau’s total prohibition.

1. The “Big Wheel” Phenomenon.

The issue most contested in the District Court can loosely be characterized as the “big wheel” phenomenon. Although the phrase seems to have meant slightly different things to different witnesses, “big wheels” can be taken broadly to refer to those inmates who exert considerable power and influence within the institution.

The thrust of the testimony by many prison administrators was that press interviews with “big wheels” tend to increase their visibility and status in the prison community. This, in turn, encourages the negative and hostile elements of the prison populace by enhancing the “big wheel’s” ability to encourage other inmates to follow disruptive paths. In the end, institutional discipline, security, and rehabilitative efforts suffer. 15

There is no question that the problems encompassed by the “big wheel” phenomenon represent legitimate Bureau concerns that, in appropriate circumstances, can justify prohibition of press interviews. We agree with the District Court, however, that the “big wheel” justification will not support a blanket ban on all interviews with all inmates.

Examination of the proceedings before the District Court reveals how broad a prohibition the Bureau has contrived for attacking this relatively isolated evil. As previously indicated, the Bureau’s Policy Statement completely bars press interviews with any inmate of any federal correctional institution, whether it be of the maximum, medium, or minimum security variety. Indeed, the Bureau’s Policy Statement prohibits press interviews with persons who are employed in business establishments or attending schools pursuant to community release programs, apparently even while they are outside the institutional boundaries.

While establishing the reach of the interview prohibition is a simple task, defining the group of inmates who might pose a threat to the interests of the penal system is not. Obviously, not all inmates represent a threat to the discipline or order of federal institutions. The testimony of Mr. Alldredge, Warden of the federal maximum security prison at Terre Haute, Indiana, indicated that only some ten percent of the inmate population in most prisons can be considered troublemakers and that, of that number, only some five percent are *1003 “most difficult.” 16 Since the “big wheel” theory is predicated on the leadership characteristics of a portion of the inmates, the potentially troublesome “big wheels” obviously must derive from the group of troublemakers and probably constitute less than the entirety of their number. Thus, although the number is admittedly imprecise, the total number of troublemaking “big wheels” is unlikely to exceed ten percent and should logically be expected to comprise a significantly smaller percentage.- 17 Moreover, while there was no testimony on this point, it would seem that the percentage of troublemakers might be expected to be smaller in the medium and minimum security institutions, and perhaps virtually nonexistent among those inmates who are allowed to participate in release work or study programs.

The Bureau interview prohibition, however, encompasses all of those persons — “big wheels,” cooperative inmates, even those who are permitted to spend a considerable portion of their time in the community at large. Absent some support to be derived from other considerations advanced by the Bureau, Policy Statement 1220.1A is too broad a prohibition to be supported by the Bureau’s “big wheel” justification. The Government has not indicated that it is unable generally to recognize troublesome “big wheels” in advance, and the evils encompassed by that phrase must be dealt with on a more selective basis. 18

2. The Federal Policy of Uniformity.

The other major justification advanced by the Bureau is that its policy of granting uniform treatment to all inmates precludes the establishment of any system that would require individual distinctions to serve as the basis for deciding whether to allow press interviews. Pointing out that the federal system constantly shifts inmates from one institution to another, the Bureau urges that any rule allowing different treatment in different institutions would incur the resentment of the inmates who are transferred into the more strictly administered facilities.

In part, this argument seems to misconstrue the order of the District Court. It appears from the testimony that much of the Bureau’s concern regarding the uniformity issue was focused on the possible discontent that would result from a transfer from an institution that imposed a fairly loose restriction to one that adhered to a tighter standard. Thus, it appears that this aspect of the *1004 Bureau’s concern rests on the seemingly erroneous assumption that the District Court order would require the Bureau to differentiate on an institutional basis by imposing one standard for interviews in a minimum security institution, another for medium security facilities, and yet a third for maximum security prisons.

This misses the thrust of the order, which is to require the Bureau to make distinctions that are based on the individualized requirements of a particular institution at a particular time, as well as on the personal attributes of the inmate seeking to participate in the interview. Compliance with the order should not generate significant discontent based on the application of vastly different standards at different institutions, for the general standard to be applied in all facilities is the same. The discontent that might be expected to arise from the transfer of inmates from one institution to another seems more likely to derive from the fact that different wardens might be expected to administer prison regulations somewhat differently. 19

To the extent that decisions affecting the denial of interviews would be influenced by a greater general concern for security in a maximum security institution, variations in press interview policy would not seem to differ in kind from any number of decisions concerning general inmate privileges. An inmate transferred into a facility with greater security needs might encounter closer confinement and greater limitations on his personal privileges. It would seem that much of his resentment of the transfer would stem from the more generalized fact of having been shifted to a more tightly administered institution. Certainly the incremental resentment that might be attributed to the specific decision to deny permission to participate in an interview does not significantly contribute to the other justifications offered in support of the absolute interview ban. Even when combined with the “big wheel” justification, this factor does not support the total ban imposed by Policy Statement 1220.1A.

The Bureau additionally asserts that the District Court order would require violation of another aspect of its policy of uniform treatment of inmates. The Bureau’s present policy is said to allow only for distinctions to be made on the basis of the “correctional needs” of the inmates. The District Court order, we are told, would require that press interview decisions be based on the “status” of the inmate seeking to participate in the interview.

Like the District Court, we find the Bureau’s distinction between “status” and “correctional need” to be somewhat opaque. If, for example, a decision to confine an inmate to a period of solitary confinement or to restrict some of his other privileges in response to repeated instances of his violent behavior is a decision based on his “correctional needs,” so might be one to deny a press request to interview him when that same conduct creates a substantial problem that the interview would create further disciplinary difficulties in the institution. 20

*1005 Thus, while we do not question that the concerns voiced by the Bureau are legitimate interests that merit protection, we must agree with the District Court that they do not, individually or in total, justify the sweeping absolute ban that the Bureau has'chosen to impose. When regulating an area in which First Amendment interests are involved, administrative officials must be careful not only to assure that they are responding to legitimate interests which are within their powers to protect; they must also take care not to cast regulations in a broad manner that unnecessarily sacrifices First Amendment rights. In this case the scope of the in-terview ban is excessive; the Bureau’s interests can and must be protected on a more selective basis. Thus, the First Amendment commands that the Bureau articulate a policy that requires the decision to grant or deny a requested interview to respond more precisely to the particular evils posed by that request. 21

Some of the Bureau’s resistance to the District Court order may stem from a misunderstanding of the scope of that ruling and the degree of latitude it provides. 22 The District Court repeatedly requested that the Bureau assist it by proposing regulations that would focus *1006 the interview policy more precisely on the problems that might arise in the individual ease. The Bureau refused, however, apparently feeling that the order permitted it no flexibility for making selective judgments. Instead, the Bureau repeatedly offered evidence of individual instances in which interviews with “big wheels” resulted in institutional disruption, notwithstanding the District Court’s insistence that its order would allow prison administrators to deny permission to interview “big wheels” whenever the interview would be likely to cause serious administrative or disciplinary problems. See note 15, supra. Finally, the Government asserted to this court that the order required that any denial of permission to conduct an interview be based on objective evidence.

This all indicates that the Bureau perhaps has read the District Court order more narrowly than was intended. The District Court’s candid observation that some of the considerations advanced by the Bureau were “somewhat impressionistic,” and its feeling that even those interests could properly be protected by the prison administrator’s exercise of discretion, suggest that decisions on requests for interviews need not be based on “objective evidence,” as the Bureau seems to fear. Courts are fully aware of the difficulties involved in making judgments in this area, and thus will be chary of impinging upon the province of the administrator. Once the warden has exercised his power “negatively on the basis of a facially legitimate and bona fide reason,” the court is not to engage in any further balancing of First Amendment considerations. Kleindienst v. Mandel, supra, at 770 of 408 U.S., at 2584 of 92 S.Ct.

The purpose and design of the District Court order, as we read it, is to require that the administrator who is closest to the situation make a considered judgment based on information concerning the individual inmate and the conditions prevailing at the institution at the time the interview is sought. What is contemplated by both the District Court and this court is not that prison administrators conduct a trial, or even an abbreviated administative hearing. It is simply that they exercise their best judgment, based on facts known to them. The District Court requirement that denials of interviews because of the disruptive actions of an individual inmate be based on his “demonstrated behavior” is not a requirement that prison administrators compile a complete dossier as a prelude to determining whether to grant an interview. It simply requires that the decisions, while sometimes grounded on “impressionistic” judgments, have a basis in the actual conduct of the inmate. To require any less would be to replace the existing closed system with one that is so subjective as to render uniformity impossible and judicial review illusory.

In sum, we think that the District Court order, properly interpreted and followed, simply requires that administrators of the federal correctional institutions make individual judgments based on their perceptions of the current requirements of their institutions, and the likely effect of a particular interview on the proper functioning of the institution. Admittedly, this will occasionally require them to make difficult decisions. But, when First Amendment rights of the press and the public weigh in the balance, such decisions are inescapable.

Only in one minor regard do we find difficulty with the District Court order. We fear that the District Court’s juxtaposition of the “probability” of serious problems and the “liklihood” of their occurrence in its definition of the conditions that would warrant a denial of a request to conduct an interview may prove somewhat confusing. Accordingly, we recast that portion of the District Court order to require that interviews be denied only where it is the judgment of the administrator directly concerned, based on either the demonstrated behavior of the inmate, or special conditions existing at the institution at the time the interview is requested, or both, that *1007 the interview presents a serious risk of administrative or disciplinary problems. The order of the District Court is affirmed as modified.

In view of the fact that it has been stayed by the Supreme Court pending this appeal, and that the Supreme Court has subsequently noted probable jurisdiction in cases generally involving the same issue, our mandate is stayed pending application to the Supreme Court for a writ of certiorari within 30 days, and thereafter as provided in Rule 41(b), Fed.R.App.P.

It is so ordered.

. Apparently order had been restored in the prisons in question. In rejecting the requests, the Director made no reference to continuing tension or unrest at those institutions.

. The ordering provisions of the District Court’s decision were cast in terms of a command that appellants should issue within 30 days new rules governing press inter-views. The rules were required to satisfy the conditions that (1) a general policy be established to permit, subject to reasonable restrictions as to time and place, confidential and uncensored press interviews with any consenting inmate, (2) exceptions to the foregoing general policy are permitted only “where it can be established as a matter of probability on the basis of actual experience that serious administrative or disciplinary problems are, in the judgment of the prison administrators directly concerned, likely to *997 be created by the interview because of either the demonstrated behavior of the inmate concerned or special conditions existing at the inmate’s institution at the particular time the interview is requested,” and (3) pending the adoption of the new regulations, press requests for interviews are to be considered on an individual basis and granted “except where it can be established that serious administrative or disciplinary problems would be created by the interview sought.”

. The regulations that govern federal visitation policy for these persons are reprinted in the appendix to the Government’s brief. Brief for the appellants on remand at 9a-15a. See also testimony of Mr. Noah H. Alldredge, Warden of the United States Penitentiary at Terre Haute, Indiana, Supp. Appendix to Original Record, Vol. I, at 216-221.

. Establishment of the inmates’ protected mailing system constituted a major change from the Bureau’s prior policy.

. Of course, these inspection tours may be suspended in case of institutional emergency. See Policy Statement 1220.1A, par. 4(b)(5). The Policy Statement in its entirety is appended to the District Court’s opinion in Washington Post I.

. See Plaintiffs’ Proposed Findings of Fact and Conclusions of Law, Findings 10-13, Supp. Appendix, Vol. I., at 14-17. These proposed findings were adopted by the District Court, Washington Post II, supra, at 784 of 357 F.Supp. In this opinion, we will use the terms “interview” and “conversation” in the manner described above.

. See Policy Statement 1220.1A, par. 8, Washington Post I, supra, at 777 of 357 F. Supp. The regulations specify that requests for exceptions can be made to the Director of the Bureau. Nothing in the record suggests that the Director’s power to make exceptions would be exercised in a manner that would significantly alter the absolute press interview ban. The District Court apparently did not feel that this power to make exceptions was a significant factor in the case, and the Government has not relied" on it in this court.

. Any First Amendment right an inmate might conceivably claim to participate in interviews with representatives of the press was not put forward as a basis for relief. However, since the position of appellees is that they should be granted general access to conduct interviews with consenting inmates, the line between the right of the press to conduct interviews and the right of the inmates to give them is not a bright one.

. See, e. g., New York Times v. Sullivan, 376 U.S. 254, 270, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964) ; Smith v. California, 361 U.S. 147, 153-154, 80 S.Ct. 215, 4 L.Ed.2d 205 (1959) ; DeJonge v. Oregon, 299 U.S. 353, 365, 57 S.Ct. 255, 81 L.Ed. 278 (1937) ; Stromberg v. California, 283 U.S. 359, 369, 51 S.Ct. 532, 75 L.Ed. 1117 (1931).

. Por example, the restrictions imposed on press access to trials are justified by the constitutional requirement that defendants be afforded a fair trial before an impartial jury, gee Estes v. Texas, 381 U.S. 532, 539-540, 85 S.Ct. 1628, 14 L.Ed.2d 543 (1965).

. In addition to the testimony of Mr. Bagdikian, who has done a great deal of prison reporting, the District Court heard the testimony of Mr. Roy M. Fisher, Dean of the School of Journalism of the University of Missouri and a former reporter and editor of the Chicago Daily News; and Mr. Timothy Leland, Assistant Managing Editor of the Boston Globe. The court received the depositions of Mr. Elie Abel, Dean of the Graduate School of Journalism of Columbia University; and Mr. John W. Machacek, news reporter for the Rochester Times-Union and winner of a 1972 Pulitzer Prize for spot news reporting for his reporting of incidents at the Attica Prison riot. Various other exhibits relating to the relationship of interviews and reporting were also put into the record.

. The proceedings in the District Oourt produced some disagreement concerning the effect of the Policy Statement’s restriction of topics of conversation to “institutional facilities programs and activities.” However, the testimony uniformly indicated that “conversations” would be considerably more limited than “interviews.”

. Mr. Liman testified that the comparisons of private interviews and group eonversa-tions at Attica revealed that the group interviews consistently produced more rhetoric than facts. Additionally, Mr. Liman testified that the inmates were reluctant to discuss certain topics in group conversation and that private interviews produced a more balanced assessment that generally was more favorable to the prison administration. Supp. Appendix, Yol. I, at 60-66.

. N.A.A.C.P. v. Button, 371 U.S. 415, 444, 83 S.Ct. 328, 344, 9 L.Ed.2d 405 (1963). In United States v. O’Brien, the Court stated that a government regulation challenged on First Amendment grounds should be upheld if it “furthers an important or substantial government interest; if the government interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest,” 391 U.S. 367, 377, 88 S. Ct. 1673, 1679, 20 L.Ed.2d 672 (1968) (Emphasis added). See also id. at 388-389, 88 S.Ct. 1673 (Harlan, X, concurring) ; United States v. Robel, 389 U.S. 258, 265-268, 88 S.Ct. 419, 19 L.Ed.2d 508 (1967) ; Shelton v. Tucker, 364 U.S. 479, 488-490, 81 S.Ct. 247, 5 L.Ed.2d 231 (1960) ; Note, The First Amendment Overbreadth Doctrine, 83 Harv. L.Rev. 844 (1970).

. There was some dispute in the evidence as to the relationship of press interviews, the “big wheel” problem, and the interests of the prison system. See the testimony of Mr. Boone, then Commissioner of the Massachusetts Department of Corrections, Supp. Appendix, Vol. I, at 126-130, and the deposition of Mr. Mattick, Professor of Criminal Justice and Director of the Center for Research in Criminal Justice at the University of Illinois, Supp. Appendix, Vol. II, at 504-516. However, the District Court determined that the problems encompassed by the “big wheel” phenomenon were “all real considerations and while somewhat impressionistic . . . supported by experience and advanced in good faith.” Washington Post I, supra, at 774 of 357 F.Supp. In the opinion of the District Court, his order allowed sufficient discretion to enable prison administrators to deal with this problem. See Supp. Appendix, Vol. I, at 131, 278.

. Testimony of Warden Alldredge, Supp. Appendix, Vol. I, at 184. The testimony of other state prison administrators tended generally to support this figure. See testimony of Mr. Boone, Supp. Appendix, Vol. I, at 139-140; testimony of Mr. Wainwright, Director of the Florida Division of Corrections, Supp. Appendix, Vol. I, at 231.

. Some problem is occasioned by the imprecision of the term “big wheel.” Some of the witnesses seemed to consider a “big wheel” to be anyone with significant status and influence in the prison community. Other testimony suggested that some prison administrators used the term “big wheel” to refer primarily to disruptive leaders. This problem does not alter our analysis. Certainly prison administrators should have no concern about publicity enhancing the prestige and influence of the leaders of the prison community who generally cooperate in maintaining order. Thus, to the extent that “big wheels” include cooperative as well as disruptive persons, the isolation of the evils presented by the “big wheel” phenomenon is further accented.

Moreover, to the extent that inmates talk to a reporter on an individual, confidential basis, giving what are essentially “not for attribution” interviews, there is no problem of enhancement of a prisoner’s status from forthcoming publicity. While prison administrators would not administer a program of prison interviews in terms of a not-for-attribution standard, the reality that reporters can and do operate on this basis is not an insignificant aspect of the total picture.

. It was asserted in oral argument that press interest is generally limited to “big wheels” and therefore that, in practical effect, the prohibition is not as broad as it might appear. The untenable premise of this argument seems to be that, since the press generally will not be interested in interviewing the average inmate, the Bureau can prohibit those interviews along with those of the troublemaking “big wheels.”

. For example, the Bureau regulation governing visitation rights of other persons contemplates individualized administration by the wardens of the various institutions, and recognizes that the facilities and requirements of each institution may require some differences in visitation rights. These differences might be expected to generate some inmate resentment, as will administration of any inmate privilege that might differ in some respects from one institution to the next.

. The Bureau previously suggested other considerations which it did not forcefully ad-vanee in the appeal to this court, but which we consider appropriate to examine. The Bureau advanced its administrative needs as one justification for the imposition of its total interview ban. This, of course, is a legitimate concern that can be accommodated by an individualized standard for determining when to grant interviews. However, this problem is no different from that posed by the administration of other visitation privileges and can be dealt with in a similar manner. If an institution is beseiged with a staggering number of requests to interview an inmate, it can legitimately devise some *1005 means of restricting press interviews to manageable proportions. For example, prison administrators might wish to impose a ceiling on the number of interviews in which an individual inmate could participate and allow the inmate himself to choose the particular reporters to whom he will grant interviews.

A second administrative concern voiced by the Bureau is the potential increase in litigation that might ensue from a system that requires it to make individual determinations. This concern is not without foundation. However, the burden should not be as monumental as the Bureau seems to think. For example, if a decision to deny permission for interviews is based on a general condition of unrest prevailing at the institution at the time, one case might dispose of the issue entirely until such time as the disruptive conditions were dissipated. If, on the other hand, denial of a series of interviews with a particular inmate was based on the threat that this individual posed to the order and security of the institution, one case likewise should dispose of the contest for a comparable period of time. In short, the Bureau should not expect that the denial of fifty requests for interviews should necessarily lead to anything approaching fifty show cause orders. More importantly, we do not feel that the incidental burden of defending the legitimacy of its decisions is a factor that merits substantial weight in our decision. Repeatedly parties complain that the “recognition” or “expansion” of constitutional rights will result in a flood of litigation by those who step forward to demand that that right be afforded them. Repeatedly, however, courts recognize that their essential duty is to vindicate those rights.

. We are aware of the Ninth Circuit decision rejecting a First Amendment challenge to these same regulations. Seattle-Tacoma Newspaper Guild v. Parker, 480 F.2d 1062 (1973). There the court concluded that the Bureau’s regulations constituted a reasonable action within the broad scope of discretion granted in the administration of maximum security institutions, but specifically refused to express any views as to the constitutionality of the application of those same regulations in federal facilities having lesser security needs. See also Hillery v. Procunier, 364 F.Supp. 196 (N.D.Cal. 1973), prob. juris, noted, 414 U.S. 1127, 94 S.Ct. 862, 38 L.Ed.2d 751 (Jan. 8, 1974) ; Pell v. Procunier, prob. juris, noted, 414 U. S. 1155, 94 S.Ct. 912, 39 L.Ed.2d 108 (Jan. 22, 1974). We note, however, that the views expressed in our opinion have found support in District Courts in the First and Fifth Circuits. See McMillan v. Carlson, 369 F.Supp. 1182 (D.Mass.1973) ; Houston Chronicle Publishing Co. v. Kleindienst, 364 F.Supp. 719 (S.D.Tex.1973).

. The part of the District Court order contemplating discretionary denials is as follows :

[Exceptions authorizing denials] shall be precisely drawn to prohibit an interview only where it can be established as a matter of probability on the basis of actual experience that serious administrative or disciplinary problems are, in the judgment of the prison administrators directly concerned, likely to be created by the interview because of either the demonstrated behavior of the inmate concerned or special conditions existing at the inmate’s institution at the particular time the interview is requested.

Washington Post I, supra, at 779 of 357 F.Supp.