Opinion for the Court filed by Circuit Judge WILKEY.
Concurring opinion filed by Circuit Judge MIKVA.
This is one of a series of cases involving disputes arising under the Freedom of In*1327formation Act (FOIA or Act) between the Central Intelligence Agency (CIA) and private parties seeking more information about the CIA’s Glomar Explorer project. The project, as many will recall from reports widely publicized in the media (though never officially confirmed), was a classified CIA program supposedly undertaken to raise a sunken Soviet submarine from the floor of the Pacific Ocean to recover the missiles, codes, and communications equipment onboard for analysis by United States military and intelligence experts.1 The project required the development of a vessel outfitted with unprecedented new technology capable of accomplishing the remarkable feat of lifting an object weighing 8.5-million pounds from a depth of 17,000-feet in the open ocean.2 The operation reportedly cost $350 million and met with uncertain success before it was cut short after its cover — as a daring venture by Howard Hughes to mine the ocean floor for manganese nodules — was blown following a mysterious burglary, in which a handful of armed men overwhelmed a guard and slipped past a sophisticated electronic alarm system, then burned their way into a Hughes safe containing a document describing the project. After the burglary, somewhat garbled information about the Glomar Explorer project somehow ended up in the Los Ange-les Times.
When news of the leak reached the CIA, Director William Colby and other officials, encouraged by the enemy’s failure during World War II to act on a Chicago Tribune story revealing that the United States had broken the Japanese naval code, scrambled to suppress further publicity about the project. They met with temporary success by briefing editors about the project in exchange for the promises of the editors not to publish accounts of the operation — at least until someone else broke the story. An impressive list of news organizations agreed to hold the story on this basis, including the New York Times, the Los Ange-les Times, the Washington Post, the Washington Star, the three major television networks, the National Public Broadcasting System, Time magazine, Newsweek, and Parade magazine. But on 18 March 1975, despite the earnest entreaties of Director Colby, columnist Jack Anderson broke the story and a spate of reports immediately appeared throughout the media.
Once the story was out, reporters and editors began to speculate as to why the CIA had undertaken the apparently hopeless task of trying to bottle up the story once it had reached the American press. Time magazine put it this way in an article published 31 March 1975:
[Tjhere is the puzzle of why so many reporters for major newspapers, magazines and TV networks simultaneously stumbled upon the [Glomar Explorer project] trail. On the morning after, some journalists got the feeling that the CIA had actually been helpful all along in getting the story out, while at the same time it apparently tried to suppress the story. There are several theories .... The last theory goes off into the wild blue yonder, suggesting that raising a Soviet submarine was not [the project’s] mission at all, but the supreme cover for a secret mission as yet safely secure.3
The appellant, Harriet Phillippi, who at the time was a Washington correspondent for Rolling Stone magazine, evidently decided to try to untangle this puzzle by using the Freedom of Information Act. She requested from the CIA “all records related to attempts by Central Intelligence Agency personnel ... to persuade any members of the news media not to broadcast, write, publish, or in any other way make public the events relating to the activities of the Glomar Explorer.”4 Following the CIA’s *1328refusal to comply with the request, the present litigation ensued.
This case is strikingly similar to a case recently decided by this court, Military Audit Project v. Casey,5 which also involved an attempt to invoke the Freedom of Information Act to acquire information about the Glomar Explorer project. In Military Audit an organization calling itself the Military Audit Project sought from the CIA documents which, among other things, would have revealed the true purpose of the Glo-mar Explorer project. We held that such information was properly withheld by the CIA under Exemption 1 to the FOIA.6
In the present case the appellant seeks documents that not only might help to reveal the purpose of the Glomar Explorer venture, but its results, if any, as well. We find the documents at issue in the present case to be exempt from disclosure under Exemption 3 of the Act for many of the same reasons which compelled our decision for the CIA in Military Audit. We therefore affirm the district court’s grant of summary judgment for the CIA.
I. PROCEDURAL HISTORY
In response to the appellant’s March 1975 request for records, the CIA refused even to confirm or deny the very existence of such records.7 The appellant, following an unsuccessful administrative appeal, then brought this action in the United States District Court for the District of Columbia. After an in camera examination of two classified affidavits submitted by the CIA, the district court held that the requested information was protected from disclosure by Exemption 3 of the Act.8 On appeal, this court reversed and remanded because the district court had resorted immediately to in camera proceedings rather than first attempting to create as complete as possible a public record of the basis on which the CIA justified its refusal to respond to the appellant’s request.9
After remand, the newly installed Carter administration abandoned the position that the CIA could neither confirm nor deny even the existence of records regarding the Glomar Explorer project; in May 1977 the government acknowledged both that the CIA was responsible for the project and that CIA officials had tried to dissuade members of the press from publishing stories about it.10 In particular, the CIA admitted the existence of 154 documents of the sort requested by the appellant. Of these, 16 were then released without deletions, 134 were released with deletions, and 4 were withheld in their entirety.11 The deletions, described as involving “sensitive details of the [Glomar Explorer] program,” 12 fell into three categories: (1) transcripts of telephone conversations between then Director of Central Intelligence William Colby and members of the press; (2) CIA descriptions of conversations between members of the press and CIA officials; and (3) CIA memoranda recounting conversations solely between CIA officials.
Unsatisfied by the CIA’s disclosures, the appellant pressed on with her litigation. On cross motions for summary judgment *1329the district court ruled for the CIA, relying on Exemption 3 of the Act, which shields from disclosure under the FOIA information specifically exempted by other statutes.13 The shielding statute invoked by the district court was section 102(d)(3) of the National Security Act of 1947, 50 U.S.C. § 403(d)(3), which provides that the “Director of Central Intelligence shall be responsible for protecting intelligence sources and methods from unauthorized disclosure.” 14 The district court expressly did not reach the question whether the information also could properly have been withheld under Exemption l,15 ruling only that under Exemption 3 the CIA need not show that the information withheld is either classifiable or classified, but only that disclosure of the withheld information “can reasonably be expected to lead to unauthorized disclosure of intelligence sources and methods ....”16 In concluding that the CIA had made the requisite showing, the district court noted that disclosure of the information sought by the appellant might reveal the purpose of the Glomar Explorer project and the extent to which its goals were accomplished,17 and that such disclosure could impair relations between the United States and the country that was the target of the project.
II. Military Audit AND THE GLOMAR EXPLORER SECRETS
The appellants in this case seek documents containing accounts and transcripts of conversations between CIA officials, members of the press, and other CIA officials regarding the purposes and results of the Glomar Explorer project. As noted above, the dispute is highly reminiscent of the litigation recently resolved by our decision in Military Audit Project v. Casey.18
In Military Audit we specifically rejected the contention that because some information about the Glomar Explorer has been leaked or officially disclosed by Government officials, all of it must be released in response to FOIA requests. We pointed out that at different times two quite different explanations of the Glomar Explorer project — both plausible — were provided to the public. The world was first told that the Glomar Explorer was designed as part of a daring project by Howard Hughes to mine the ocean seabed for manganese nodules; this story was widely believed. Later, however, the story changed and the world was informed that the purpose of the vessel actually was to raise a stricken Soviet submarine from the ocean floor. This explanation has also been widely accepted. As we pointed out in Military Audit:
What should be obvious is that if it is both plausible that the Glomar Explorer was designed to mine the seabed and at the same time also plausible that the Glo-mar Explorer was designed to raise a Russian submarine, it is plausible that the Glomar Explorer was in fact designed to perform yet some still-secret third function. Someday, when the story is safe to tell, we may discover, in the words of Time Magazine, “that raising a Soviet submarine was not [the Glomar Explorers] mission at all, but the supreme cover for a secret mission as yet safely secure.” And even if the true purpose of the mission was in fact to raise a submarine from the floor of the ocean, there may be some advantage in leaving the Soviet in*1330telligence agencies with lingering doubts whether some other purpose motivated the project. Whatever the truth may be, it remains either unrevealed or unconfirmed.19
If in fact the Glomar Explorer was designed to perform some still undisclosed function, then the project had not just one but two cover stories. The first — the story about finding manganese nodules — was an initial innocent cover. It quite satisfactorily explained various activities connected with the project which could not possibly be kept secret, such as the construction and deployment of the vessel. But, in view of the size of the undertaking, involving as it did the effort of a great many individuals and business and Government entities, the CIA would have been only prudent to create a fallback cover story to be used in the event the initial facade cracked. Such a story would have to be plausible, and, if it were successfully to satisfy the public, it would have to involve a covert undertaking seemingly of great importance. The story that the Glomar Explorer project was designed to lift a Soviet submarine from the bottom of the ocean appears to fit the bill.
Almost as important as the substance of the fallback cover would be the way in which the CIA protected it. Any appearance of carelessness would tend to undercut the credibility of the fallback story. The CIA would, therefore, be required to appear to devote as much zeal to protecting the fallback story from disclosure as it would devote to shielding the truth from revelation. Paradoxically, however, when the initial cover story was blown, it would be important for the CIA to arrange for the rapid and convincing dissemination of the fallback story in order to prevent inquisitive reporters from stumbling onto the truth before they could be sold on the authenticity of the fallback story.
If there was a fallback cover story for the Glomar Explorer project — that the vessel was designed to raise a sunken Russian submarine from the ocean floor — Director Colby’s apparently simultaneous efforts to cover up the fallback story while at the same time assuring its widespread dissemination begin to make sense. And so does the Government’s sudden switch in position once the fallback story appeared throughout the media. For once the fallback story was out, the Government suddenly had to reverse its position and begin refusing to confirm or to deny the stories it could have been planting in the press only days earlier; if the Government openly “admitted” the fallback story, it would thereby in effect declassify it, and as a result lose its ability to refuse to open its files on the subject.
Moreover, if the CIA from time to time relies on fallback stories, in order to maintain this option it would be helpful for the CIA to treat alike all revelations about secret programs which might possibly involve a fallback story. Thus, even if the truth is out about some mission, the CIA might profit by treating the truth in the same way it would handle a fallback story. For if it were to do otherwise, the CIA would risk telegraphing to its adversaries the authenticity or inauthenticity of the explanations publicly given for intelligence projects whose initial, innocent cover had been blown.
In sum, the line between what may be revealed and what must be concealed is itself capable of conveying information to foreign intelligence agencies. For this reason, this court cannot simply assume, over the well-documented and specific affidavits of the CIA to the contrary, that revelation of seemingly innocent information which might nonetheless jeopardize a fallback cover story is required under the FOIA, either because the information in question has already been made public, or even, as in the present case, because it was disseminated for confidential purposes by the CIA itself. Without the ability to engineer controlled leaks of disinformation, the CIA would be deprived of the ability to disseminate a fallback cover while simultaneously protecting it.
Note that actions by the CIA and its officials become inherently ambiguous with the idea of a fallback cover story in mind. *1331For example, in the present case the appellant makes much20 of the apparently inadvertent “revelations” in the French edition of William Colby’s autobiography in which Colby spends several pages describing the Glomar Explorer project, its purported results, and his efforts to keep the Glomar Explorer story bottled up.21 The CIA is said to have expurgated from the English edition of the book all references to the purpose and results of the mission, but supposedly Colby’s publisher failed to pass the required deletions on to the French publisher, thereby permitting the unexpurgated story to slip out. From this episode the appellant concludes that the truth is out.
The appellant may be right. It may be that these revelations simply reflect lack of coordination among a former Director of Central Intelligence and his publishers. Or they may reflect Colby’s personal judgment that the Glomar Explorer story may now be told, a judgment he mistakenly believed would be shared by the current CIA bureaucracy. On the other hand, this story is also understandable as a modest effort by the CIA and a former Director to buttress the credibility of a widely disseminated fallback cover story. At the least, this latter explanation is consistent with everything in the record describing Colby’s handling and explanations of the incident.
In sum, as we found in Military Audit, the “cat is not out of the bag.” There may be much left to hide, and if there is not, that itself may be worth hiding.
III. THE DOCUMENTS WITHHELD
Three classes of documents have been withheld from the appellant: (1) transcripts of conversations between Director Colby and members of the media; (2) CIA documents describing contacts between CIA officials and members of the media regarding the Glomar Explorer, and (3) documents describing internal communications between officials of the CIA regarding the CIA’s effort to dissuade the American press from publishing stories regarding the Glo-mar Explorer mission.
The third category of documents just described is clearly covered by the holding in Military Audit. Like the documents in dispute in that case, internal documents between CIA officials regarding the purposes and results of the Glomar Explorer mission plainly could reveal information that would lead to the disclosure of intelligence sources and methods. The district court properly granted summary judgment to the CIA with respect to these deletions. We affirm the district court with respect to these documents without further ado.
We now turn to the other withheld documents, those involving transcripts or memoranda relating to contacts between CIA officials and the press. With respect to these categories of information, the appellant argues that the revelation of interchanges between the CIA and the press could not be of any assistance to a foreign adversary attempting to get at the real truth of the matter without some guarantee that what the CIA told the press was in fact true.
This theory has some superficial appeal, but on reflection its allure, though seductive, cannot withstand inspection. First, there is obviously the possibility that the American press did not publish everything disclosed by the CIA at its confidential briefings. William Colby’s entreaties may have been at least partially successful. If so, this case appears in much the same posture it would have had if Colby had been entirely successful. For had Colby been successful, a journalist who, like the appellant here, had not been briefed, might have filed an FOIA request similar to the one which the appellant litigates here. Such a journalist could have alleged — as the appellant has in the present case — that he merely sought to be treated on an equal basis with the other journalists who had been given confidential information about the project. He could have pointed out — as the appellant has in the present case — that the CIA had provided the information at issue to *1332other journalists who lacked security clearances and who did not promise to respect the secrecy of the information provided to them. But such a hypothetical FOIA request would surely fail, for Exemption 3 could be successfully invoked by the CIA to bar release of the documents requested.
Neither we nor the appellant knows for sure that everything the CIA disclosed has in fact been printed. We do know that the published accounts of the results of the project differ over the critical question whether all, or a part, of the Soviet submarine was recovered.22 If Director Colby told some journalists about the outcome of the project, but those journalists voluntarily and patriotically abstained from publishing that information, disclosure of the documents requested by the appellant could lead a foreign intelligence analyst to information they would otherwise not have obtained.
Furthermore, without the disclosure of the documents demanded by the appellant, foreign analysts remain in the dark as to the provenience of the information appearing in published reports. Some of it may have come from Director Colby and other CIA officials, but no one who was not privy to the CIA disclosures can know for sure which information came from CIA sources and which information originated elsewhere — unless the appellant receives the documents she requests. Release of those documents would thus not be as innocuous as the appellant would have us believe.
There may be other ways in which a foreign analyst might be able to capitalize on the information the appellant seeks. The affidavits submitted to the district court on which the district court based its grant of summary judgment plainly indicate with requisite specificity the concerns of the CIA that disclosure of this information would threaten the revelation of intelligence sources and methods.23 These affidavits are entitled to substantial weight, which the district court properly gave them. We do not believe the appellant has succeeded in overcoming their weight merely by asserting that the content of the discussions between CIA officials and members of the press cannot jeopardize national security because the CIA officials in question may have been lying to the media. Indeed, they may have been. But the FOIA does not require the CIA to lighten the task of our adversaries around the world by providing them with documentary assistance from which to piece together the truth.
Finally, of course, there is the possibility that the Glomar Explorer project really was designed to raise a sunken Russian submarine, that the project was in fact successful, and that the information obtained compromised Soviet security. If so, public confirmation of the results of the project might publicly humiliate the Soviets. As Director Colby pointed out in his own autobiography,24 Khrushchev stated in his Memoirs that what led him to cancel the Paris Summit meeting with President Eisenhower after the U-2 incident was not the fact that American U-2’s had overflown the Soviet Union — that was not news to Khrushchev— but rather that President Eisenhower had publicly admitted that he had approved the mission. The parallel to the Glomar Explorer project is obvious. In the world of international diplomacy, where face-saving *1333may often be as important as substance, official confirmation of the purpose of the Glomar Explorer project through release of transcripts and memoranda detailing the conversations of the highest CIA officials could have an adverse effect on our relations with the Soviets. The affidavits submitted to the district court by the CIA indicate the real possibility that such a negative reaction could occur. The appellant has not provided any evidence to the contrary.
For these reasons, the district court was correct in granting summary judgment for the CIA.25
Affirmed.