delivered the opinion of the Court.

This case presents the question whether state prisoners who have commenced habeas corpus proceedings in a federal district court may, in proper circumstances, utilize the instrument of interrogatories for discovery purposes.

I.

Petitioner is the Chief Judge of the United States District Court for the Northern District of California. Respondent is the warden of the California State Prison at San Quentin. The proceeding was initiated by Alfred Walker who had been convicted in the California courts of the crime of possession of marihuana. After exhaust*289ing state remedies, he filed a petition for habeas corpus in the Federal District Court, alleging that evidence seized in the search incident to his arrest was improperly admitted at his trial. The basis for this claim was his allegation that the arrest and incidental search were based solely on the statement of an informant who, according to Walker’s sworn statement, was not shown to have been reliable; who, in fact, was unreliable; and whose statements were accepted by the police without proper precautionary procedures.

The District Court issued an order to show cause and respondent made return. Thereafter, Walker filed a motion for an evidentiary hearing, which the District Court granted. Two months later, Walker served upon the respondent warden a series of interrogatories, pursuant to Rule 33 of the Federal Rules of Civil Procedure, seeking discovery of certain facts directed to proof of the informant’s unreliability. Respondent filed objections to the interrogatories, alleging the absence of authority for their issuance. The District Judge, without stating his reasons, disallowed the objections and directed that the interrogatories be answered. Respondent applied to the Court of Appeals for the Ninth Circuit for a writ of mandamus or prohibition. The Ninth Circuit vacated the order of the District Court. It held that the discovery provisions of the Federal Rules of Civil Procedure were not applicable to habeas corpus proceedings and that 28 U. S. C. § 2246, the statutory provision specifically relating to the use of interrogatories in habeas corpus proceedings, did not authorize their use for discovery. Wilson v. Harris, 378 F. 2d 141 (1967).

Because of the importance of the questions presented and the diversity of views among the district and appellate courts that have considered the problem,1 we granted *290certiorari. 392 U. S. 925. We agree with the Ninth Circuit that Rule 33 of the Federal Rules of Civil Procedure is not applicable to habeas corpus proceedings and that 28 U. S. C. § 2246 does not authorize interrogatories except in limited circumstances not applicable to this case; but we conclude that, in appropriate circumstances, a district court, confronted by a petition for habeas corpus which establishes a prima facie case for relief, may use or authorize the use of suitable discovery procedures, including interrogatories, reasonably fashioned to elicit facts necessary to help the court to “dispose of the matter as law and justice require.” 28 U. S. C. § 2243. Accordingly, we reverse and remand the case in order that the District Court may reconsider the matter before it in light of our opinion and judgment.

II.

The writ of habeas corpus is the fundamental instrument for safeguarding individual freedom against arbi*291trary and lawless state action. Its pre-eminent role is recognized by the admonition in the Constitution that: “The Privilege of the Writ of Habeas Corpus shall not be suspended . . . U. S. Const., Art. I, § 9, cl. 2. The scope and flexibility of the writ — its capacity to reach all manner of illegal detention — its ability to cut through barriers of form and procedural mazes — have always been emphasized and jealously guarded by courts and lawmakers. The very nature of the writ demands that it be administered with the initiative and flexibility essential to insure that miscarriages of justice within its reach are surfaced and corrected.

As Blackstone phrased it, habeas corpus is “the great and efficacious writ, in all manner of illegal confinement.” 2 As this Court said in Fay v. Noia, 372 U. S. 391, 401-402 (1963), the office of the writ is “to provide a prompt and efficacious remedy for whatever society deems to be intolerable restraints.” See Peyton v. Rowe, 391 U. S. 54, 65-67 (1968).

It is now established beyond the reach of reasonable dispute that the federal courts not only may grant evi-dentiary hearings to applicants, but must do so upon an appropriate showing. Townsend v. Sain, 372 U. S. 293, 313 (1963); Brown v. Allen, 344 U. S. 443, 464, n. 19 (1953). And this Court has emphasized, taking into account the office of the writ and the fact that the petitioner, being in custody, is usually handicapped in developing the evidence needed to support in necessary detail the facts alleged in his petition, that a habeas *292corpus proceeding must not be allowed to founder in a “procedural morass.” Price v. Johnston, 334 U. S. 266, 269 (1948).

There is no higher duty of a court, under our constitutional system, than the careful processing and adjudication of petitions for writs of habeas corpus, for it is in such proceedings that a person in custody charges that error, neglect, or evil purpose has resulted in his unlawful confinement and that he is deprived of his freedom contrary to law. This Court has insistently said that the power of the federal courts to conduct inquiry in habeas corpus is equal to the responsibility which the writ involves: “The language of Congress, the history of the writ, the decisions of this Court, all make clear that the power of inquiry on federal habeas corpus is plenary.” Townsend v. Sain, supra, at 312.

In the present case, we are confronted with a procedural problem which tests the reality of these great principles. We are asked by Walker to establish the existence of rights for those in custody to discover facts which may aid their petitions for release. We are asked to do this by declaring that the provisions of the Federal Rules of Civil Procedure granting such rights to litigants in civil causes are available to Walker; or if we refuse so to conclude, to affirm the existence of power in the District Court to authorize discovery by written interrogatories. We address ourselves to those issues.

III.

Rule 1 of the Federal Rules of Civil Procedure provides that: “These rules govern the procedure in the United States district courts in all suits of a civil nature . . . with the exceptions stated in Rule 81.” At the time of the decision below Rule 81 (a) (2) provided, in relevant part, that the Rules were not applicable in habeas corpus “except to the extent that the practice in *293such proceedings is not set forth in statutes of the United States and has heretofore conformed to the practice in actions at law or suits in equity.”3

The Court of Appeals for the Ninth Circuit held that the second requirement — “conformity” with practice— made it necessary to show that “prior to September 16, 1938, discovery was actually being used in habeas proceedings, and that such use conformed to the then discovery practice in actions at law or suits in equity.” 378 F. 2d, at 144. No such showing was made and it is not here contended that it can be made. Walker contends, however, that the rule requires only a showing that habeas proceedings conformed generally to pre-existing practice in law and equity, and he contends that this general requirement is met.

We need not consider this contention that the Court of Appeals took an unnecessarily restricted view of the thrust of the “conformity” requirement, because for other reasons we conclude that the intended scope of the Federal Rules of Civil Procedure and the history of habeas corpus procedure, make it clear that Rule 81 (a) (2) must be read to exclude the application of Rule 33 in habeas corpus proceedings.

It is, of course, true that habeas corpus proceedings are characterized as “civil.” See, e. g.. Fisher v. Baker, 203 U. S. 174, 181 (1906). But the label is gross and *294inexact.4 Essentially, the proceeding is unique. Habeas corpus practice in the federal courts has conformed with civil practice only in a general sense. There is no indication that with respect to pretrial proceedings for the development of evidence, habeas corpus practice had conformed to the practice at law or in equity “to the extent” that the application of rules newly developed in 1938 to govern discovery in “civil” cases should apply in order to avoid a divergence in practice which had theretofore been substantially uniform. Although there is little direct evidence, relevant to the present problem, of the purpose of the “conformity” provision of Rule 81 (a)(2), the concern of the draftsmen, as a general matter, seems to have been to provide for the continuing applicability of the “civil” rules in their new form to those areas of practice in habeas corpus and other enumerated proceedings in which the “specified” proceedings had theretofore utilized the modes of civil practice. Otherwise, those proceedings were to be considered outside of the scope of the rules without prejudice, of course, to the use of particular rules by analogy or otherwise, where appropriate.5

*295Such specific evidence as there is with respect to the intent of the draftsmen of the rules indicates nothing more than a general and nonspecific understanding that the rules would have very limited application to habeas corpus proceedings. At the very least, it is clear that there was no intention to extend to habeas corpus, as a matter of right, the broad discovery provisions which, even in ordinary civil litigation, were “one of the most significant innovations” of the new rules. Hickman v. Taylor, 329 U. S. 495, 500 (1947). Walker does not claim that there was any general discovery practice in habeas corpus proceedings prior to adoption of the Federal Rules of Civil Procedure.

In considering the intended application of the new rules to habeas corpus, it is illuminating to note that in 1938 the expansion of federal habeas corpus to its present scope was only in its early stages. Mooney v. Holohan, 294 U. S. 103 (1935): Johnson v. Zerbst, 304 U. S. 458 (1938); Waley v. Johnston, 316 U. S. 101 (1942). It was not until many years later that the federal courts considering a habeas corpus petition were held to be required in many cases to make an independent determination of the factual basis of claims that state convictions had violated the petitioner’s federal constitutional rights. Brown v. Allen, 344 U. S. 443 (1953); Townsend v. Sain, 372 U. S. 293 (1963). In these circumstances it is readily understandable that, as indicated by the language and the scanty contemporary exegesis of Rule 81 (a) (2) which is available, the drafts*296men of the rule did not contemplate that the discovery-provisions of the rules would be applicable to habeas corpus proceedings.

It is also of some relevance that in 1948, when Congress enacted 28 U. S'. C. § 2246 expressly referring to the right of parties in habeas corpus proceedings to propound written interrogatories, its legislation was limited to interrogatories for the purpose of obtaining evidence from affiants where affidavits were admitted in evidence. Again, the restricted scope of this legislation indicates that the adoption in 1938 of the Federal Rules of Civil Procedure was not intended to make available in habeas corpus proceedings the discovery provisions of those rules.

Indeed, it is difficult to believe that the draftsmen of the Rules or Congress would have applied the discovery rules without modification to habeas corpus proceedings because their specific provisions are ill-suited to the special problems and character of such proceedings. For example, Rule 33, which Walker here invoked, provides for written interrogatories to be served by any party upon any “adverse party.” As the present case illustrates, this would usually mean that the prisoner’s interrogatories must be directed to the warden although the warden would be unable to answer from personal knowledge questions relating to petitioner’s arrest and trial. Presumably the warden could solicit answers from the appropriate officials and reply “under oath,” as the rule requires; but the warden is clearly not the kind of “adverse party” contemplated by the discovery rules, and the result of their literal application would be to invoke a procedure which is circuitous, burdensome, and time consuming.

The scope of interrogatories which may be served under Rule 33 also indicates the unsuitability of applying to habeas corpus provisions which were drafted without reference to its peculiar problems.

*297By reference to Rule 26 (b), the rule would give the prisoner a right to inquire into “any matter, not privileged, which is relevant to the subject matter involved in the pending action,” whether admissible at trial or not. This rule has been generously construed to provide a great deal of latitude for discovery. See Hickman v. Taylor, supra, at 507; 2A Barron & Holtzoff, supra, § 646. Such a broad-ranging preliminary inquiry is neither necessary nor appropriate in the context of a habeas corpus proceeding.

Except for interrogatories to be served by the “plaintiff” within 10 days after the commencement of “the action,” Rule 33 provides that the interrogatories may be served without leave of court. The “adverse party” must then take the initiative to contest the interrogatories and a hearing in court on his objections is required. Unavoidably, unless there is a measure of responsibility in the originator of the proceeding, the “plaintiff” or petitioner, this procedure can be exceedingly burdensome and vexatious. The interrogatory procedure would be available to the prisoners themselves since most habeas petitions are prepared and filed by prisoners, generally without the guidance or restraint of members of the bar. For this reason, too, we conclude that the literal application of Rule 33 to habeas corpus proceedings would do violence to the efficient and effective administration of the Great Writ. The burden upon courts, prison officials, prosecutors, and police, which is necessarily and properly incident to the processing and adjudication of habeas corpus proceedings, would be vastly increased; and the benefit to prisoners would be counterbalanced by the delay which the elaborate discovery procedures would necessarily entail.

It is true that the availability of Rule 33 would provide prisoners with an instrument of discovery which could be activated on their own initiative, without prior *298court approval, and that this would be of considerable tactical advantage to them in the prosecution of their efforts to demonstrate such error in their trial as would result in their release. But despite the forceful and ingenious argument of Walker’s counsel and amici curiae,6 this consideration cannot carry the day. It is a long march from this contention to a conclusion that the discovery provisions of the Federal Rules of Civil Procedure were intended to extend to habeas corpus proceedings. We have no power to rewrite the Rules by judicial interpretations. We have no power to decide that Rule 33 applies to habeas corpus proceedings unless, on conventional principles of statutory construction, we can properly conclude that the literal language or the intended effect of the Rules indicates that this was within the purpose of the draftsmen or the congressional understanding.

IV.

To conclude that the Federal Rules’ discovery provisions do not apply completely and automatically by virtue of Rule 81 (a)(2) is not to say that there is no way in which a district court may, in an appropriate case, arrange for procedures which will allow development, for purposes of the hearing, of the facts relevant to disposition of a habeas corpus petition. Petitioners in habeas corpus proceedings, as the Congress and this Court have emphasized, and as we have discussed, supra, at 290-292, are entitled to careful consideration and plenary processing of their claims including full opportunity for presentation of the relevant facts. Congress has provided that once a petition for a writ of habeas corpus is filed, unless *299the court is of the opinion that the petitioner is not entitled to an order to show cause, the writ must be awarded “forthwith,” or an order to show cause must be issued. 28 U. S. C. § 2243. Thereafter, if the .court concludes that the petitioner is entitled to an evidentiary-hearing, cf. Townsend v. Sain, supra; 28 U. S. C. § 2254, it shall order one to be held promptly. 28 U. S. C. § 2243.

Flexible provision is made for taking evidence by oral testimony, by deposition, or upon affidavit and written interrogatory. 28 U. S. C. § 2246. Cf. §§ 2245, 2254 (e). The court shall “summarily hear and determine the facts, and dispose of the matter as law and justice require.” 28 U. S. C. § 2243. But with respect to methods for securing facts where necessary to accomplish the objective of the proceedings Congress has been largely silent. Clearly, in these circumstances, the habeas corpus jurisdiction and the duty to exercise it being present, the courts may fashion appropriate modes of procedure, by analogy to existing rules or otherwise in conformity with judicial usage. Where their duties require it, this is the inescapable obligation of the courts. Their authority is expressly confirmed in the All Writs Act, 28 U. S. C. § 1651. This statute has served since its inclusion, in substance, in the original Judiciary Act as a “legislatively approved source of procedural instruments designed to achieve ‘the rational ends of law.' ” Price v. Johnston, 334 U. S. 266, 282 (1948), quoting Adams v. United States ex rel. McCann, 317 U. S. 269, 273 (1942). It has been recognized that the courts may rely upon this statute in issuing orders appropriate to assist them in conducting factual inquiries. American Lithographic Co. v. Werckmeister, 221 U. S. 603,609 (1911) (subpoenas duces tecum); Bethlehem Shipbuilding Corp. v. NLRB, 120 F. 2d 126, 127 (C. A. 1st Cir. 1941) (order that certain documents be produced for the purpose of pretrial discovery). In Price v. Johnston, supra, this Court held *300explicitly that the purpose and function of the All Writs Act to supply the courts with the instruments needed to perform their duty, as prescribed by the Congress and the Constitution, provided only that such instruments are “agreeable” to the usages and principles of law, extend to habeas corpus proceedings.

At any time in the proceedings, when the court considers that it is necessary to do so in order that a fair and meaningful evidentiary hearing may be held so that the court may properly “dispose of the matter as law and justice require,” either on its own motion or upon cause shown by the petitioner, it may issue such writs and take or authorize such proceedings with respect to development, before or in conjunction with the hearing of the facts relevant to the claims advanced by the parties, as may be “necessary or appropriate in aid of [its jurisdiction] . . . and agreeable to the usages and principles of law.” 28 U. S. C. § 1651.

We do not assume that courts in the exercise of their discretion will pursue or authorize pursuit of all allegations presented to them. We are aware that confinement sometimes induces fantasy which has its basis in the paranoia of prison rather than in fact. But where specific allegations before the court show reason to believe that the petitioner may, if the facts are fully developed, be able to demonstrate that he is confined illegally and is therefore entitled to relief, it is the duty of the court to provide the necessary facilities and procedures for an adequate inquiry. Obviously, in exercising this power, the court may utilize familiar procedures, as appropriate, whether these are found in the civil or criminal rules or elsewhere in the “usages and principles of law.”7

*301Accordingly, we reverse the judgment of the Court of Appeals for the Ninth Circuit and remand the case for further proceedings in accordance with this opinion.

Reversed and remanded.

dissenting.

I would affirm the Court of Appeals’ holding in this case, Wilson v. Harris, 378 F. 2d 141, that 28 U. S. C. § 2246 does not authorize discovery in habeas corpus proceedings. Upon affirmance I would not go further and write what appears to me to be in effect an advisory opinion directing the trial court to formulate some kind of new legal system for discovery in this kind of case. Fully agreeing with the Court’s statement that “[w]e have no power to rewrite the Rules by judicial interpretations,” I go further and doubt that we have power to *302direct lower courts to write new laws providing for discovery in habeas corpus cases. This is a complicated field of law making and I think we should not enter this field in the absence of some valid delegation of legislative power by the Congress. Since I cannot agree that. Congress has granted us such power, I am unable to go along with the Court’s opinion.

There have been many complaints among members of the bar about many Court-made rules of procedure and I would venture the suggestion that in no field have the number of those complaints exceeded the complaints in this particular field of discovery. I regret that I cannot “assume,” with the Court, that given blanket authority, “courts in the exercise of their discretion will [not] pursue or authorize pursuit of all allegations presented to them.” This case makes me skeptical about such an assumption. Here Walker was convicted in a state court of having marihuana in his possession. After exhausting all state remedies he asked the federal courts to let him out of jail. He apparently did not allege his innocence, does not now do so, and this Court apparently does not now consider the question of guilt or innocence in this case. What he does allege is that the trial court made an error in admitting certain evidence against him. It is not alleged that the evidence was not relevant against him or that the verdict resting on that evidence was not a truthful, honest verdict. We must, therefore, assume that he was and is guilty of the crime of which he was convicted. See my dissent in Kaufman v. United States, ante, p. 231, decided today. What is relevant, however, and all that is alleged, is that the evidence used against him, presumably the marihuana, was found on his premises as the result of a search made after a statement by a person to a policeman, which statement the allegations now charge “was not shown to have been reliable” and which was *303made by a person “who was in fact unreliable.” It may be possible that a new trial over this issue can establish that the person telling the officer that marihuana could be found on Walker’s premises was an “unreliable” person and that the statement he made was also “unreliable.” But the fact remains that the marihuana was found where the unreliable person’s unreliable statement told the officer it would be found. Consequently it appears to me that the present case against a defendant whose guilt has been proved to a jury beyond a reasonable doubt should not be taken as an appropriate one on which this Court lays the groundwork for a new and vast judicial legislative rule-making program.

Perhaps it might not be considered amiss mildly to suggest that in cases like this, where records contain no question at all about guilt, some convictions should at some time be treated as final and no longer subject to challenge, at least by collateral attack. Although I admit that Aguilar v. Texas, 378 U. S. 108 (1964), Spinelli v. United States, 393 U. S. 410 (1969), and other recent cases go a long way, I had not previously thought that even these cases could support what the Court is doing in this case.

dissenting.

I agree that neither Rule 33 of the Federal Rules of Civil Procedure nor any statute authorizes the interrogatories sought in this case. I further agree that district courts do have power to require discovery when essential to render a habeas corpus proceeding effective. But I would make it explicit that such power is narrow and should be exercised sparingly, and would not set the district courts “at large,” as I fear today’s opinion may be taken to do.

*304I.

This case furnishes an apt illustration of the differences between my viewpoint and what seems to be that of the majority. As stated more fully in the Court's opinion, ante, at 288-289, Walker claimed that marihuana admitted at his trial was seized incident to an arrest which was based upon information supplied by an unreliable informant. After the District Court had ordered an evi-dentiary hearing, Walker directed to the respondent warden a series of interrogatories designed to establish the unreliability of the informant. The interrogatories asked whether the officer who arrested Walker had made previous arrests or searches on the basis of information given by the same informant; if so, whether such arrests or searches resulted in convictions; and whether the informant had ever supplied information which the officer considered unreliable.

It seems apparent that this discovery was not essential to an adequate habeas proceeding. All of the information sought was known to the arresting officer. Walker knew the officer’s identity; in fact, the officer had testified at the preliminary hearing and at trial on the very issue of the informant’s reliability. Hence, there is no reason to believe that all of the information could not have been obtained by calling the officer as a witness at the habeas hearing. Although I realize that the parties have not directed their arguments to this precise question, I am satisfied that on the face of things Walker cannot possibly show that this discovery is essential to a fair proceeding. Accordingly, I would affirm outright the judgment of the Court of Appeals.1

*305II.

The more troublesome aspect of the Court’s opinion is its long-run implications. For it can be taken as suggesting that the best solution to the problem of discovery in habeas corpus proceedings is to permit each district court to devise “appropriate modes of procedure” on a case-by-case basis. As regards the immediate future, a case-by-case approach may be unavoidable, since there is at present no body of applicable discovery rules and the district courts must have power to order discovery which is essential to effective disposition of habeas applications. But I consider that from a broader standpoint the problem of habeas discovery should be dealt with not case by case but through exercise of our rule-making power. See 28 U. S. C. § 2072.

There are several reasons for believing that a case-by-case approach will be unsatisfactory in the long range. It seems to me that in fairness both to habeas petitioners and to their adversaries, the discovery procedures which are available in such actions should be uniform throughout the federal system and not dependent upon the vary*306ing “discovery attitudes” of particular district judges. If discovery procedures are developed case by case, there will at the least be a very long period during which procedures will differ from district to district. Even assuming that a coherent body of rules finally will emerge because of the unifying influence of appellate decisions, it is unlikely that the rules thus generated will be the best that could have been devised. Appellate courts, including this one, are imperfectly informed both about the extent of the need for additional discovery in habeas corpus and about the procedures best suited to meet those needs and to achieve prompt dispatch of habeas proceedings. They are, therefore, poorly situated to lay down guidelines for the district courts. Moreover, discovery rules fashioned in the course of day-to-day adjudication are likely to suffer from the limitations which accompany that process.

Such considerations lead me to think that, in the longer view, the formulation of discovery rules can best be accomplished through use of the power which Congress has conferred upon us to establish general rules governing civil procedure in the federal district courts. By using this method of rule making, the advice of the Judicial Conference of the United States and its appropriate advisory committees could be obtained.2 These bodies are well equipped to assess the dimensions of the discovery problem and devise apt solutions. Their deliberations would be free from the time pressures and piecemeal character of case-by-case adjudication. And the resulting rules would be uniform throughout the federal system.

*307My conviction that this would be the best course is strengthened by recollection of our decision in Miner v. Atlass, 363 U. S. 641 (1960), and the events which followed. In Miner we held that a District Court sitting in admiralty had no power to order the taking of an oral discovery deposition. Responding to a suggestion in our opinion, see 363 U. S., at 651, and to earlier stirrings at the bar, the Judicial Conference and the Advisory Committee on General Admiralty Rules swiftly proposed new Admiralty Rules authorizing certain additional kinds of discovery, including oral depositions. After approval by this Court and submission to Congress, as required by statute,3 the new Admiralty Rules went into effect a little more than a year after our decision.4 There is no reason to think that the Judicial Conference and the advisory committees would not be equally cooperative in this instance.

For the reasons stated in Part I of this opinion, I would affirm the judgment of the Court of Appeals in this case.

dissenting.

I concur with most of what is said in the Court’s opinion, but cannot concur in its judgment.

I wholly agree that Rule 33 is inapplicable to habeas corpus proceedings. Contrary to my Brother Harlan, I further agree that federal judges in carrying out their duty to dispose of habeas corpus applications “as law and justice require,” 28 U. S. C. § 2243, should not be inhibited by inflexibly formalized procedural rules. In getting at the facts in such cases, the district courts *308should be given wide leeway for “discretion to exercise their common sense.” Machibroda v. United States, 368 U. S. 487, 495.*

However, for the reasons stated in Mr. Justice Harlan’s dissenting opinion today in Kaufman v. United States, ante, p. 242, which I have joined, I would affirm the judgment in the present case.