OPINION OF THE COURT
ADAMS, Circuit Judge.This appeal, in an action challenging the parole guidelines promulgated by the United States Parole Commission, raises two issues of broad import. First, it requires us to examine the conditions under which a class action, which the trial court refused to certify, may be submitted to an appellate court despite the fact that the named plaintiff no longer retains a “live” personal grievance. Second, it presents the question of the validity, under both statutory and constitutional standards, of the guidelines that govern federal grants of parole.
I. THE FACTS
A. The Guidelines
Beginning in 1910, certain prisoners incarcerated for conviction of federal crimes have been eligible for release on parole.1 To facilitate such arrangement, the United States, in 1948, established a Parole Board (the Board), within the Department of Justice, to rule on applications for parole. While originally the Board’s decisions were not based on formally articulated policies and procedures, in 1973 the Board published a series of regulations governing parole decisions, including “guidelines” to establish “customary release dates” for given classes of offenders.2
In 1976, Congress enacted the Parole Commission and Reorganization Act (the PCRA).3 The PCRA reconstituted the Parole Board as the United States Parole Commission (the Commission), an independent federal agency. Under the PCRA, the Commission is responsible for promulgating “guidelines” for the exercise of statutory discretion concerning the granting of parole.4 In making a decision regarding an individual inmate, the Commission is directed to examine “the nature and circumstances of the offense and the history and characteristics of the prisoner,” and then to determine whether release would “depreciate the seriousness of his offense,” “promote disrespect for the law” or “jeopardize the public welfare.” Inmates are to be *242released “pursuant to the guidelines” if the determination by the Commission, in light of the statutory criteria, is favorable.5
The “guidelines” currently utilized are substantially the same ones that channeled the Board’s discretion before the enactment of the PCRA. Under them, offenses are assigned a “severity” rating, and are placed into one of six categories, ranging from “low” to “greatest.” Each inmate is assigned a “parole prognosis score” of between 0 and 11, using “salient factors” such as the age at which the inmate was first convicted, his employment background, his drug history, his previous parole revocations, and his prior convictions. The “guidelines” include a grid in which a combination of salient factor score and offense severity rating identifies a “customary” time span to be served.6
B. The Named Plaintiff
John M. Geraghty, a Chicago policeman, was convicted in 1973 of conspiracy to commit extortion and of making false declarations to the grand jury. The extortion charge was based on Geraghty’s use of his position as a Chicago Vice Squad Sergeant to shake down local dispensers of alcoholic beverages; the false declaration charge arose out of denials of involvement in this activity. After attempts to overturn his sentence proved unsuccessful,7 Geraghty applied for parole. Despite the good institutional adjustment one might expect from a former policeman, parole was denied Ger-aghty on the ground that:
Your offense has been rated as very high severity. You have a salient factor score of 11. You have been in custody for a total of 4 months. Guidelines established by the Board for adult cases which consider the above factors indicate a range of 26-36 months to be served before release for cases with good institutional program performance and adjustment. After review of all relevant factors and information presented, it is found that a decision at this consideration outside the guidelines does not appear warranted.
As finally amended, Geraghty’s sentence was 30 months. Thus, under the “customary release date,” he could not be granted parole before the end of his sentence as reduced by “good time” credits.
*243Geraghty applied for parole a second time, but that request was also denied in June, 1976. The second statement of reasons given by the Commission was substantially identical with the first.8 Asserting that he was being denied parole by reason of a mechanical application of the guidelines, Geraghty then brought the present suit as a class action challenging the validity of the guidelines, and questioning the procedures by which the guidelines were applied to his case.9
Judge R. Dixon Herman, of the District Court for the Middle District of Pennsylvania, declined to certify the class action, and granted summary judgment against Ger-aghty on all of the claims he had asserted. After an appeal from both rulings was docketed, but before oral argument was heard by us, Geraghty’s sentence expired and he was released.
II. JURISDICTIONAL PROBLEMS
Before proceeding to the merits of Ger-aghty’s contentions, we must consider a number of significant procedural objections.
A. Jurisdictional Basis of Suit
Initially, the question of the jurisdictional underpinning of the suit before us must be resolved. In his complaint, Geraghty claimed jurisdiction under (1) 28 U.S.C. § 2241 (the habeas corpus statute), (2) 5 U.S.C. §§ 700-706 (the Administrative Procedure Act), and (3) 28 U.S.C. § 1331 (federal question jurisdiction). The trial court held that under Preiser v. Rodriguez,10 habeas corpus is the only remedial base available to the plaintiff. To the contrary, however, we conclude that it is appropriate to treat this action as one for declaratory judgment under 5 U.S.C. §§ 700-706 (1970) and 18 U.S.C. § 4218(c) (1976).
In Preiser, state prisoners brought suit seeking an injunction restoring “good time” credits that they claimed were unconstitutionally taken from them. The Supreme Court held that since the prisoners were challenging the fact or duration of their imprisonments, § 1983 was unavailable, and their sole federal remedy was by habeas corpus. The Preiser opinion rested on two grounds. First, the Court noted that the interest in federal-state comity weighed against the advisability of allowing state prisoners to bypass the exhaustion requirement 11 of the state habeas corpus statute.12 Second, the Court stated that the more specific provisions of the habeas corpus act should be read to modify the general cause of action granted by § 1983.13
Neither of these considerations is applicable here. The courts face no barriers resulting from federal-state relations in adjudicating issues such as the ones before us, since the present controversy involves the application of a federal statute by federal authorities. And, unlike a habeas corpus action challenging state confinements, no exhaustion has been statutorily mandated. Indeed, in contrast to the situation in Preiser, Congress expressly contemplated declaratory actions to challenge the provisions of the federal parole guidelines. 18 U.S.C. § 4218(c) (1976) declares that Parole Commission actions, except for individual parole decisions, are to be reviewable under the Administrative Procedure Act.14 The legislative history of § 4218(c) states, inter alia:
*244This section brings the Commission rule-making process within the coverage of the Administrative Procedure Act judicial review procedures. In this regard, the Conferees recognize the principles established in Pickus v. United States, 165 U.S.App.D.C. 284, 507 F.2d 1107 (1974).15
Pickus v. Parole Board,16 entailed a challenge by prisoners to parole guidelines brought as an action for declaratory judgment under the APA. It appears that Congress in citing Pickus clearly evinced an intent to allow suits like the one before us to proceed by way of an action for declaratory judgment.17
Moreover, even under the strictures of Preiser, itself, the present action would not be inexorably channeled into the form of a habeas corpus proceeding. While the relief requested for Geraghty included “enlargement from custody” pending review of his parole status, such request is now moot. In comparison, the class relief sought was (1) a declaration that the parole guidelines are invalid, and (2) an injunction against further actions denying parole to other federal prisoners on the basis of the guidelines. This relief falls within the Supreme Court’s holding in Wolff v. McDonnell, 418 U.S. 539, 554-55, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974), that Preiser does not bar either a declaratory judgment or a prospective injunction against enforcement of unconstitutional regulations relating to revocation of good time credits. The class does not demand that its members be released on parole, but only that the Parole Board not utilize the guidelines in evaluating future parole applications.
Therefore we conclude that this suit may proceed as an action for declaratory judgment.18
B. Mootness
We are next faced with a challenge to this Court’s jurisdiction on the basis of mootness. After the appeal was filed, but prior to oral argument, Geraghty was released from confinement. Since, at trial, Judge Herman declined to certify this case as a class action, the government contends that the matter is now moot because of Geraghty’s release. The argument proceeds that Geraghty, as the named plaintiff, has no further stake in the operation of the parole guidelines, and no class, in fact, has been certified to assert an interest in the guidelines.19
Geraghty makes two responses to the claim of mootness. First, he declares that challenges to the parole guidelines represent a situation where a legal injury is “capable of repetition yet evading review.” Prisoners will often be released before their cases are finally resolved in the appellate courts. Therefore, he avers, the case before *245us comes within a traditional exception to the mootness doctrine.20 Second, Geraghty maintains that his case is encompassed by the principle that a class action is not moot simply because the claims of the named plaintiff are rendered academic. He urges that the failure of Judge Herman to grant class action status was an abuse of discretion and that subsequently granted class status should be permitted to “relate back” to the time when Geraghty had a live claim.21
Resolution of the justiciability of this case in light of the mootness objection is best undertaken in the context of a somewhat extended discussion of this evolving doctrine.22
1. Contours of Mootness Doctrine
The principle that a court may not decide a moot case arose primarily from rules of equity and common law, combined with traditional notions of the functions of courts.23 It was only in the last decade that the attitude of the Supreme Court toward the adjudication of moot cases became definitively intertwined with the mandate of Article III, which provides that the power of the judiciary is limited to cases or controversies.24
Still more recently, however, the Supreme Court has made clear that elements of the “mootness” doctrine find their roots not only in constitutional dictates, but also in more flexible considerations of policy.25 The first step in our mootness analysis, *246therefore, must be to attempt to etch the outlines of the constitutional elements of this doctrine.
The constitutional command that matters submitted to the judiciary for resolution must be within the category of “case or controversy”
limits the business of the federal courts to questions presented in an adversary context, and in a form historically viewed as capable of resolution through the judicial process. And in part those words define the role assigned to the judiciary in a tripartite allocation of power to assure that the federal courts will not intrude into areas committed to the other branches of government.26
As we understand the constitutional requirements, a case presented for adjudication must be an actual, concrete dispute over legal rights; the controversy may not be a hypothetical one.27 In addition, at the commencement of suit, the dispute must concern some individual plaintiff who is injured by the wrong in question.28
However, once a suit meeting these conditions has been instituted, the limitations of Article III do not absolutely require that an individual who is personally harmed by the wrongs continue to have a live dispute. Rather, the issues with regard to jurisdiction are:
(1) Whether a legal controversy exists sufficient to establish that the case is not hypothetical.
(2) Whether the controversy affects an individual in a concrete case sufficient to provide the factual predicate for the reasoned adjudication which is the province of the judiciary.
(3) In addition, the court must answer the more policy-oriented question whether the parties before it have, at the time for decision, sufficient functional adversity to sharpen the issues for judicial resolution.
The existence of a plaintiff with standing at the outset of the litigation insures the initial fulfillment of these three conditions; but the elimination of the individual grievance may bring such fulfillment into question. Once suit has been commenced, it nonetheless remains open to the court to determine that a case or controversy persists despite the disappearance of the named plaintiff as a “live” litigant.29 Moreover, since the third consideration is not an absolute Article III requirement, in evaluating the degree of functional adversity, the court may ascribe weight to reasons of policy.30
We reach these conclusions on the basis not only of the Supreme Court’s explications of the mootness doctrine, but in re*247liance on cases in which the Supreme Court has actually exercised jurisdiction.
2. Actions Surviving Loss of Claim by the Named Plaintiff
(a) Repetitious Evasion of Review
The first genre of disputes in which the Supreme Court has not required a continuing live stake was described by our Court in United States v. Frumento as the “most traditional of exceptions to the mootness doctrine.”31 This group of cases has utilized the rubric “capable of repetition yet evading review” to characterize controversies whose effect on plaintiffs is, by their very nature, of limited duration. In such instances it is unlikely that any person will retain a live stake in the outcome of the case by the time it has made its way through the appellate process. Nonetheless, the actions complained of are likely to be repeated, and therefore present the possibility of a recurring but a judicially irremediable wrong unless jurisdiction is retained.32
In many of these cases, the party who originally brought suit can present a plausible argument that he or she will be subject to a recurrence of the conduct at issue, and accordingly continues to have a personal stake in the controversy. To this extent, the plaintiff retains a “live” interest. Such hypotheses, however, in many cases would clearly fail to meet the standards imposed on litigants at the outset of a case to show “standing,”33 and in others seem to be keyed to the notion that the conduct at issue may be repeated with respect to someone within the class represented, although not with respect to the particular named plaintiff.34
If the disputed action will, with reasonable probability, recur in the near future and the issues may be resolved without reference to nuances of particular fact situations, the constitutional prerequisites of a legal controversy and a concrete factual predicate may be satisfied despite the lack of a named plaintiff before the court who retains his original interest in the conflict. The potential for repetition with respect to such plaintiff may itself be one indication of the satisfaction of the further prudential criterion of functional adversity, but the unfairness of an evasion of review may also *248be relevant to the exercise of discretion. Repetition is not, however, the sole possible indication of either constitutional or discretionary justiciability.
(b) Class Actions
In the related area of the mootness of class actions, the Supreme Court has consistently held that once a class action is certified, the mooting of a claim set forth by the named plaintiff does not automatically deprive a court of jurisdiction over the cause of action asserted by the class. A justiciable legal controversy may continue to exist between the class as an entity and the defendant, thus satisfying Article III. Originally, this holding was articulated in the context of class actions dealing with claims that were also capable of repetition but evading review.35
But the Supreme Court has, in other situations, sustained the justiciability of what may be referred to as “headless” class actions; that is class actions in which the named plaintiff retains no “live” claim. In Richardson v. Ramirez,36 the Court adjudicated the complaint of a class composed of California ex-prisoners, despite the fact that the three named plaintiffs already had obtained the relief they sought. The action challenged the denial of voting rights to ex-felons under California law. By the time the case reached the appellate court, however, the three named ex-felons had already been registered to vote by the named defendants, three county clerks charged with enforcing election statutes. Nonetheless, the California Supreme Court issued a declaratory judgment.
On review, the Supreme Court noted that inasmuch as the opinion of the California tribunal was binding on all county clerks, the underlying issue evaded review by being “incapable of repetition.”37 In addition, since the Court interpreted the case as an action against a class composed of all county clerks, it identified a “live” controversy between the non-consenting clerks and felons who, in the future, might wish to register in their districts. Although it is clear that the named plaintiffs could not have commenced suit simply by claiming that they wished to move to the other counties and register to vote,38 once suit was appropriately instituted, the Supreme Court took notice of an on-going controversy with unnamed class members.
More recently, in Franks v. Bowman Transportation Co.,39 an action under Title VII of the Civil Rights Act of 1964, the Supreme Court acknowledged the viability of a headless class action, where the lone named plaintiff had lost his eligibility for relief as a result of subsequent misconduct. Inasmuch as the plaintiff class had been properly certified, the Court held that a continuing legal controversy was clear, and the case was not constitutionally moot. The Court also attributed the earlier linkage of headless class actions to situations capable of repetition but evading review to “policy rules” rather than to the constitutional element of justiciability.40 Identifying a situation repetitiously evading review was only one method of establishing sufficient “functional adversity,” the Court held. In Bowman the class certification combined with the factors that the class members were easily identifiable, clearly entitled to a remedy, and had demonstrated competence and tenacity in their litigation subsequent to the disqualification of the class repre*249sentative, satisfied the requirements of the mootness doctrine.41
From the Ramirez and Bowman cases, plus the class-oriented “repetition” cases, we draw the conclusion that a continuing conflict between a clearly-defined class and a defendant remains an Article III case or controversy even after the named plaintiff loses his personal stake in the outcome. And, if the evidence indicates a possibility of evasion of review, or a clear continuing functional adversity, the discretionary component of the mootness doctrine is also satisfied.
Thus, in the ease before us, if Geraghty’s suit had been properly certified as a class action before he had been released from prison, it would have been jurisdictionally appropriate to continue to entertain the suit even after his release. Obviously, a large number of prisoners remain subject to the parole guidelines, and insofar as the guidelines result in their being denied parole, such prisoners clearly are adversely affected. In adjudicating this matter, a court therefore would not be concerned merely with an abstract or hypothetical conflict. And a proper class certification would bring into focus the “functional adversity” necessary for adjudication.
3. The Impact of Denial of Class Certification
In this case the trial court refused class certification. The question, then, is whether such refusal combined with Geraghty’s release prevents our review of this matter, on the ground that we have before us neither a “live” plaintiff nor a properly certified class.
The Parole Commission earnestly contends that under Board of School Commissioners of Indianapolis v. Jacobs,*42 in the absence of a properly certified class the mooting of a named plaintiff automatically removes the suit from the category of a justiciable case or controversy. Reading Jacobs in this fashion, however, is incompatible with the Supreme Court’s holdings in both prior and subsequent cases.43
In Gerstein v. Pugh,44 the short duration of challenged pretrial custody precluded a class from being certified before the named plaintiffs were released. Nevertheless, the district court certified a class of pretrial detainees, and entertained the action. Noting that the challenged procedures would continue to apply to a succession of pretrial detainees, the Supreme Court held that the class certification preserved justiciability even though the certification might have been entered after the named plaintiffs lost a “live” interest in the suit.45
*250The Supreme Court in Baxter v. Palmigi-ano46 adjudicated a controversy concerning prison regulations, although the case had not been properly certified as a class action and the named plaintiffs had died or had been released prior to the oral argument on appeal. The Supreme Court held that the subsequent intervention of another prisoner brought a justiciable case before the Court. This was so despite the fact that it was not clear that when the intervention had occurred, the district court still had a “live” individual plaintiff before it.
Most recently, in United Airlines v. McDonald,47 the Supreme Court reviewed a refusal to allow putative class members in an employment discrimination suit to intervene in order to appeal a denial of class certification. By the time the motion for intervention was presented, the original named plaintiffs had already been awarded relief and as to them the case was moot. Again, intervention was permitted at a time when the named plaintiffs no longer retained “live” claims.
In light of Gerstein, handed down on the same day as Jacobs, and of Baxter and McDonald, both decided after Jacobs, the Jacobs case should not be viewed as standing for the proposition that the federal courts are constitutionally barred from continuing to adjudicate disputes when the named plaintiff no longer retains his claim. So long as a factually concrete legal controversy continues to exist, it would appear that the constitutional power of a court over the case remains. Consequently, the holding of Jacobs is perhaps best understood as a specific instance that must be seen in the context of the over-arching set of principles adumbrated by the authorities cited above.
In Jacobs, the Supreme Court reviewed a First Amendment challenge by a group of students to regulations limiting publication of a newspaper. Since the students had graduated, and the periodical in question had ceased publication, the inability to delimit accurately the plaintiff class deprived the Court of the assurance of justiciability. In addition, the failure of counsel to seek a proper class certification cast serious doubt on the continued adequacy of the adversarial clash.48
An appropriately defined and certified class assures the courts that the case involves a concrete legal controversy, proper factual predicate, and functional adversity. The lack of certification, however, does not inevitably require dismissal, if the *251elements of justiciability are otherwise established.49 Thus, in Gerstein, Baxter, and McDonald no doubt could validly have been raised concerning the continuation of the underlying legal controversies. Counsel in Gerstein and McDonald, could not have been charged with responsibility for the failure of certification, and in both Baxter and McDonald the continuation of the adversarial difference was assured by the intervention of live plaintiffs. In those situations, a concrete dispute continued and representation was adequate. Consequently, the cases were not dismissed.
In the. matter before us, there is neither a contention by the government nor evidence in the record which raises any question but that a legal controversy exists. The fate of numerous federal prisoners continues to turn in large part on the application of the guidelines in question. Such a situation in our view satisfies the “case or controversy” components of the mootness doctrine. Further, under the circumstances of this case, the more discretionary elements of the restrictions against entertaining moot cases do not mandate dismissal. Four aspects of the case lead to this conclusion.
First, Geraghty’s action, while not wholly congruent, shares many characteristics with the cases denominated “capable of repetition, yet evading review.” This is so since federal prisoners are eligible for parole if their sentences exceed one year.50 Accordingly, while some prisoners will retain their grievances long enough to achieve appellate review, a number of prisoners with short sentences, like Geraghty, will inevitably be discharged before they háve an opportunity to litigate fully the legality of the rules which deny them parole. This alone is a factor weighing heavily in favor of justicia-bility.51 And since it is the prisoners with unusually short sentences to whom the parole guidelines deny the benefit of the trial judge’s leniency,52 the limited probability of review for a prisoner with a short sentence is particularly pertinent.
Second, as we have noted, this case involves denial of a class action certification. As Chief Judge Seitz observed in Gardner v. Westinghouse,53 if mooting of a named plaintiff’s claim bars review of a denial of class certification, our rule against interlocutory appeals of class certification orders will, in a significant number of cases, effectively immunize from review such adverse class determinations. This is hardly a salu*252tary result for, an improper denial of a class certification could bar adjudication of an otherwise constitutionally justiciable controversy.
Third, the attorneys for Geraghty, while not possessed of a legally continuing relationship with members of the plaintiff class,54 have nonetheless undertaken this litigation on a class-oriented basis. There is no indication of any diminution of vigor in their efforts despite the release of Ger-aghty. Indeed, as already observed, they represent another individual plaintiff who now seeks to intervene in the matter. Consequently, there is a prima facie case of functional adversity, a central element which the mootness doctrine seeks to preserve.
Finally, the major issues in the case in no way appear to be tied to the nuances of individual fact patterns. The discharge of Geraghty does not alter either the interests of the members of the putative class55 or the practice of the Parole Commission in applying its guidelines.56
Hence, if class certification is appropriate in this case, the mootness of Geraghty’s claim should not bar adjudication.
C. Trial Court’s Refusal to Certify Class Action Status
Class certification was denied by the district court as neither “necessary nor appropriate.”57 It was not necessary, Judge Herman held, because the possibility of avoiding mootness is not expressly comprehended in the criteria of Rule 23. We agree with the trial court that a possibility of avoiding mootness on appeal would not, of itself, be a sufficient basis for conferring class action status on a suit otherwise barred by Rule 23. Rule 23, however, allows litigants to bring class actions so long as they meet its standards. The plaintiff here need not have proved that certification was “necessary,” as the trial judge seemed to indicate, but only that there was compliance with the prerequisites of Rule 23.
Judge Herman stated that the class action was not “appropriate” on a number of grounds. First, he correctly held that two of the issues raised — namely, the classification of Geraghty’s offense under the guidelines and Geraghty’s access to certain Commission files — had no class-wide applicability. Second, the trial judge held that since the challenge to the guidelines based on the conflict with the statute under which Ger-aghty was sentenced,58 “is inapplicable to all members of the proposed class,” class certification as to all prisoners was inappropriate. Finally, the court noted that while the guidelines had the effect of lengthening Geraghty’s incarceration, they would also have the effect of shortening the length of confinement of individuals who had been sentenced to terms greater than three times the period recommended by the guidelines. Thus, the court concluded, Geraghty’s claims were not typical of the class he purported to represent.59
The trial judge is correct in his observation that not all of the grounds of action alleged in the complaint are applicable to the class of “all federal prisoners who have been or will become eligible for release on parole.” 60 The conclusion that this implies that a class action is inappropriate, however, does not properly acknowledge the powers and duties of the trial court *253under section (c)(4) of Rule 23.61 Under section (c)(4), the trial judge has the power to certify certain issues as subject to class adjudication, and to limit overbroad classes by the use of sub-classes. Indeed, this authority may be exercised sua sponte.62
Failure to exercise such power in a proper case has been held to be an abuse of discretion. In Samuel v. University of Pittsburgh,63 for example, we concluded that the trial court’s decision to decertify a class because of the administrative difficulty of computing damages was an abuse of discretion. If managerial difficulties were present; we held, “some investigation into the possible usefulness of subclasses as suggested by Rule 24(c)(4)(B), should have been undertaken before decertification was ordered.” 64 •
Consequently, the district court need not have refused class certification in toto because certain claims were inapplicable to the entire class. Rather, with respect to those claims, it could have certified the class as to the prisoners to whom the claims applied. It was completely open to the trial judge to refuse certification with respect to the claims personal to Geraghty (access to files and offense severity classification), but then to certify claims relating to 18 U.S.C. § 4208(a) (1970) only as to prisoners sentenced under that statute, and to certify the broader statutory and constitutional claims as to the entire prisoner class. A forbearance to consider these options constituted a failure properly to exercise discretion.
Similarly, the refusal to certify entirely because of potential inconsistencies between Geraghty’s interest and those of other members of the putative class was improper in this case. First, it is not clear that a divergence in interest exists. It is true that prisoners who have been paroled under the guidelines have no legal interest in challenging the guidelines. Since, however, invalidation of the guidelines would not nullify their paroles, they have no interest adverse to the plaintiffs.
On the other hand, it could be argued that prisoners whose parole dates have been scheduled, under the guidelines, for the future have an interest in maintaining the assurance provided by their presumptive parole dates. However, for prisoners like Geraghty, whose “customary release dates” fall beyond the length of time for which they are imprisoned, the guidelines are of no possible benefit. Within this last subclass, at least, there is no incompatibility of *254interest. By not considering the use of Rule 23(c)(4) to establish a subclass, the court did not exercise an informed discretion and therefore its action cannot be sustained.65
Accordingly, we reverse the denial of class certification. However, since it is appropriate that the district court make the initial evaluation of the proper subclasses, the ease should be remanded on this point.66
III. THE VALIDITY OF THE GUIDELINES
A remand for resolution of the class certification dispute would improvidently dissipate judicial effort if the district court were correct in its determination that Geraghty’s substantive contentions are devoid of merit. Thus, it is fitting, and indeed necessary, for us to consider the merits of Geraghty’s claim regarding the guidelines.
Geraghty offers three major challenges to the validity of the guidelines: (1) the guidelines’ “fixed and mechanical” approach violates both the PCRA and the Constitution; (2) the guidelines transgress both statutory and constitutional commands insofar as they fail to take account of the sentences imposed by the district courts; and (3) as applied to prisoners sentenced before their adoption, the guidelines constitute impermissible ex post facto legislation.67
A. The PCRA and the Guidelines
The first two contentions require analysis of the statutory scheme of the PCRA. Since this is true, and since the government maintains that the PCRA specifically ratified the practices in question, we shall analyze the propriety of the alleged “fixed” and “mechanical” nature of the guidelines together with their alleged disregard of judicially imposed sentences.
1. The Characteristics of The Guidelines
Before beginning our exploration of the PCRA, we outline the features of the guidelines to which Geraghty objects. According to Geraghty, the guidelines find their origin in decisions by the Parole Board to categorize a series of 51 “offenses” into six severity levels, — low, low moderate, moderate, high, very high, and greatest — without regard to the actual sentences imposed for each offense.68 For the various severity levels, the median length of time served by prisoners in each of three “prognosis categories”69 became the three “customary release dates” for that level. While admitting that in fixing severity levels it essentially followed the procedure which Ger-aghty outlined, the Commission denies that the median incarceration formed the basis of the “customary release date.” Ger-*255aghty’s proof is sufficient to raise a material issue of fact as to the method by which the “customary release date” is calculated by the Commission.
Geraghty alleges that, after the Commission has set its grid of “customary release dates,” such release dates predetermine the time when prisoners will be paroled without regard to the individual facts of each case. This assertion is given substance by the Commission’s admission that only 8.7% of the parolees in 1975 were released before the “customary release date,” and by the apparent practice of referring only to the guidelines in denials of parole.70 The Commission responds that under its regulations all “relevant evidence” is taken into account,71 that decisions outside of the guidelines are permitted where circumstances warrant,72 and that in fact guidelines represent “objective standards by which the ad hoc judgment of an individual’s offense severity is measured.”73 However, the Commission presented no affidavits as to the actual working of parole decision-making. And on this subject, there appears to be a genuine difference regarding material facts.
Equally important, Geraghty notes that the sentence actually imposed by the trial judge is not a factor in determining the customary release date, and avers that under current procedures the sentence is not given any weight in the individual parole-determination process. He observes that current regulations have removed the prescribed sentence from the considerations which must be taken into account. Significantly, the Commission, in its trial brief, admitted that no weight is given to the length of the sentence imposed by the trial judge.74
Since this case comes before us from a dismissal by summary judgment, and since Geraghty has provided factual support for his characterization of the guidelines, we must take his account as correct for purposes of this appeal.75
2. The Legislation
Although in 1973, the Parole Board had promulgated guidelines substantially similar to those currently being questioned, the first legislative authorization for such guidelines was contained in the PCRA of 1976. As part of a legislative overhaul of the parole system, the 1976 Act granted the newly-established Parole Commission the power and the duty to:
promulgate rules and regulations establishing guidelines for the power [to grant or deny parole].
18 U.S.C. § 4203(a)(1) (1976).
These guidelines are to be used in making parole decisions in accordance with statutory criteria:
(a) If an eligible prisoner has substantially observed the rules of the institution or institutions to which he has been confined, and if the Commission, upon con-*256sidefation of the nature and circumstances of the offense and the history and characteristics of the prisoner, determines;
(1) that release would not depreciate the seriousness of his offense or promote disrespect for the law; and
(2) that release would not jeopardize the public welfare; subject to the provisions of subsections (b) and (c) of this section, and pursuant to guidelines promulgated by the Commission pursuant to section 4203(a)(1), such prisoner shall be released.
18 U.S.C. § 4206 (1976).
Geraghty points to the statutory mandate that the parole decision be based on the “nature and circumstances of the offense and the history and characteristics of the prisoner.” He maintains that under the guidelines, as they are utilized by the Commission, prisoners are denied individualized consideration of the “nature and circumstances” of their offenses, a consideration directed by the statute.
The Commission replies that nothing in the statute explicitly constrains its discretion in this regard. It declares that in the first instance, the “nature and circumstances” are evaluated in the course of assigning a “severity rating” and, in any event, the provision in its regulations for decisions outside the guidelines satisfies any further statutory requirements.
Moreover, the Commission observes that while the statute explicitly requires consideration of a number of factors,76 the length of the sentence imposed is not among them. Geraghty answers that since the control by the trial judge over the sentencing process is an integral feature of the criminal justice system as currently structured, the sentence actually imposed is “additional relevant information” whose consideration by the Commission must necessarily have been contemplated by Congress.
The text of the statute, in our opinion, seems capable of supporting either interpretation advanced. We therefore turn to other lines of analysis.
3. The Legislative History
Geraghty contends that, before the guidelines came into effect, the length of sentence imposed by the trial judge — rather than the Parole Commission’s evaluation of the “severity” of the offense — constituted the primary means of establishing the length of imprisonment necessary to vindicate society’s interest in incarceration.77 Whether or not this is accurate, under H.R. 5727, the direct ancestor of the PCRA,78 the *257sentence imposed by the trial judge was of controlling significance. Under H.R. 5727, a federal prisoner with good institutional behavior was to have been released upon completion of one-third of his sentence, unless the Commission affirmatively found release to be undesirable under the statutory criteria.79 The virtue of having a “definite parole date at one-third of the sentence”80 was the aspect of the bill which prompted the greatest comment in floor debate81 and no remarks by any Congressman indicate that the Commission was to be responsible for re-evaluating the adequacy of sentences.
There is no implication in the legislative history of the House Bill that the approach of the guidelines in establishing classes of offenders based on the Parole Commission’s views of the severity rating of “offenses” in general was to have been ratified by the legislation. Indeed, comments during the floor debates manifest hostility to such an approach.82
In contrast, the Senate version of the legislation anticipated the adoption of guidelines identical with those then in existence under the Parole Board.83 These would guide the Commission in deciding whether the prerequisites for parole exist: “a reasonable probability that such person will remain at liberty, without violating the law,” and “release is not incompatible with the welfare of society.”84 The decision was to be made substantially on the basis of “a report by the proper institutional officer,” and the Commission’s investigation.85
Further, in contrast to the House Bill, the Senate Bill specifically contemplated the new Commission’s responsibility as encompassing revision of inappropriate judicially imposed sentences. The introduction to the Senate Report described parole as “an ex*258tension of the sentencing process.”86 The Report also referred to the “benefits” of the guidelines in “reducing the opportunity for sentencing disparity” and ensuring that “offenders sentenced for similar crime under similar circumstances will be required to serve comparable periods of incarceration” regardless of the statute under which they are sentenced.87 Such a function would, of course, not be possible if the length of sentence imposed itself entered into the parole decision.
*257“(a). A prisoner shall be released on parole if his record shows that he has substantially observed the rules of the institution in which he is confined on the date of his eligibility for parole, unless it is determined by an examining panel (as provided in section 4207(a)) that he should not be released on such date for one or more of the following reasons: (1) there is a reasonable probability that such prisoner will not live and remain at liberty without violating any criminal law. (2) there is a reasonable probability that such release would be incompatible with the welfare of society; or (3) the prisoner’s release on such date would so depreciate the seriousness of his crime as to undermine respect for law.
*258In the words of Congressman Kastenmeir, a sponsor of the House version, the bill which ultimately emerged from the Conference “required considerable compromise on the part of conferees on both sides.” 88 The controlling effect of the sentence length was eliminated, but on the other hand characterization of parole as an “extension of the sentencing process” was deleted from the statement of purposes otherwise identical with the Senate version.89 The Bill is described as “having the practical effect of balancing differences in sentencing policies between judges and courts,” rather than having the purpose of eliminating sentencing disparities.90
Unlike the Senate version, the legislative history of the conference bill did not contain explicit endorsement of the guidelines as they were currently promulgated,91 and the Commission was directed to be “cognizant of past criticism of parole decision making”.92 Moreover, a provision for judicial review of the guidelines themselves was adopted from the House version.
In describing the parole eligibility criteria, the Conference Report stated:
[I]t is the intent of the conferees that the Parole Commission review and consider both the nature and circumstances of the offense and the history and characteristics of the prisoner . . . the Parole Commission, in making each parole determination shall recognize and make a determination as to the relative severity of the prospective parolee’s offense . . .93
The parole decision makers, apparently in each case,
must weigh the concepts of general and special deterrence, retribution and punishment, all of which are matters of judgment . and come up with determinations of what is' meant by “would not depreciate the seriousness of the offense or promote disrespect for the law” that, to the extent possible, are not inconsistent with findings of other parole decisions.94
As an example of this process, the conferees postulate a situation in which a public official is convicted of fraud and sentenced to three years imprisonment. In such a case, states the Report, “his release on parole after one year might satisfy the depreciate the seriousness criterion, but the Commission could justify denying release on *259grounds that such release would promote disrespect for the law.”95
If Geraghty’s description of the effect given the parole guidelines by the Commission is accurate, the Commission in effect is following the views of the Senate version of the PCRA, rather than the policies of the Conference Committee.
According to Geraghty’s description, the Commission attempts to eliminate sentence disparity by completely ignoring the length of sentence imposed by the trial judge.96 This policy would fail to acknowledge that under the Conference version, a leveling of sentence disparity is a by-product of other policies, rather than a goal. In its hypothetical the Conference Committee apparently saw the length of sentence as being a relevant factor. To fail to take account of the length of sentence imposed is thus at odds with the contemplated effect of the legislation. The fact that the prePCRA guidelines ignored sentence length is not dispositive since the accompanying regulations at that time explicitly mandated that length of sentence be taken into account in considering deviation from the guidelines.97
Similarly, Geraghty alleges that the guidelines were derived by a grouping of offenses into severity levels by the Commission and by subsequently averaging the time customarily served by prisoners in each “level,” and that these levels are given controlling effect by the Commission to the exclusion of other factors. Such a process seems significantly different from the view expressed by the conferees that in structuring “each parole determination [the Commission] shall make a determination as to the relative severity of the prospective parolees offense.”98 Nor is it consistent with the procedure contemplated in the Committee’s hypothetical. And the refusal to incorporate the previous individualized determination by the trial judge would exacerbate this difficulty.
Thus, if Geraghty’s recapitulation of the function and genesis of the guidelines is supported by the evidence, there are important divergences between the Parole Commission’s actions and the intent of Congress in enacting the statutory mandate.
4. The Question of Constitutionality
The differences between Congressional intent and the alleged operation of the guidelines also must be examined against the backdrop of the disparities between the procedures Geraghty describes and the structure of the balance of the criminal justice system. To the extent that the Parole Commission makes individual judgments about the relative culpability of prisoners and the length of imprisonment proper to vindicate the needs of society yet fails to take account of the sentence imposed by the court, the Commission embarks, alone, on a task which is the traditional province of the judiciary. Insofar as the Commission attempts to make general rules as to the appropriate punishment for crimes which effectively bind parole decisions in all cases, it undertakes functions which are usually discharged by the legislature. Yet the Commission lacks the institutional safeguards of either the courts or the Congress: it has neither judicial independence and allegiance to the ideal of dispassionate and reasoned decisions nor legislative responsiveness to the popular will.
Such tensions and the accompanying constitutional doubts that they raise need not be resolved or cast a shadow on the validity of the PCRA, of course, unless Congress has clearly authorized the assumption by the Commission of judicial and leg*260islative functions.99 In light of the legislative history discussed above, we find no such clear authorization in the PCRA.
(a) Tensions with the Judiciary
Federal criminal statutes generally provide a range of penalties from which the sentencing court must choose. In making such choices the federal courts draw on the perception that “For the determination of sentences, justice generally requires consideration of more than the particular act by which the crime was committed, and that there be taken into account the characteristics and propensities of the individual.”100 The Supreme Court has recognized that “[t]he belief no longer prevails that every offense in a like legal category calls for an identical punishment without regard to the past life and habits of a particular offender.” 101
District courts may exercise their discretion in individual cases largely untrammeled by appellate review.102 They are not free, however, to eschew the task they are called upon to perform by adopting rules to fix given punishments on given classes of offenders. Such a position by the sentencing judge would constitute an abuse of discretion calling for a vacation of the sentence imposed.103
If the PCRA provided for the guidelines to function as Geraghty alleges, a federal defendant would find himself in the anomalous situation of being sentenced by a judge who is forbidden to utilize fixed penalties, but having parole granted by a Commission that is required to use such penalties and which, at the same time, refuses to take account of the decision of the sentencing court. The rigid categories called for by the parole guidelines would frequently nullify the discretion which the trial judges are required to exercise in evaluating each case on its individual merits.
It may well be that Congress itself could foreclose all sentencing options.104 But, absent a clear statement of such a purpose, we should be wary of attributing to Congress the intent to allow the Parole Commission to undertake such a wide-ranging overhaul of the criminal sentencing process.
This is particularly so in view of the fact that the Commission has disclaimed *261any intention of giving rehabilitation a major role in its parole decisions.105 Parole has traditionally been conceived of as “an established variation on imprisonment of convicted criminals [whose] purpose is to help individuals reintegrate into society as constructive individuals as soon as they are able.”106 For such a purpose, the Commission, which can observe the prisoner during his incarceration, has a significant advantage over the court in determining eligibility for parole under a standard which is distinct from the traditional role of the judiciary.107 When, however, the parole authority focuses consideration entirely on factors of deterrence, incapacitation and retribution, it takes into account almost exclusively the very factors that are available to the sentencing judge. The Commission then begins to perform functions which are within the traditional province of the judiciary.108 At least where the prior determination of the judicial branch are given no weight, therefore, serious questions are raised whether the constitutional protections provided by an independent judiciary are being undermined.109 Since Congress has manifested no clear intention to raise such questions, we do not interpret the PCRA as authorizing such disregard of judicial sentences.110
(b) Delegation of Legislative Functions
The determination of the proper range of punishment which may be levied *262in regard to a class of crimes, like the determination of what actions are criminal, is within the ambit of the legislative prerogative.111 “Because of the seriousness of criminal penalties,” they should in general attach only after the body responsible for forming our laws in the first instance has established that such penalties are appropriate.112 Criminal sanctions serve an amalgam of social goals, and their severity is best determined by a body which can respond to the balance of such goals as prescribed by the popular will.113
If the parole guidelines function as automatically as Geraghty alleges that they do, the Commission is effectively redrafting the penalty provisions of the United States criminal code.114 The severity levels of the guidelines neither “fill in the details” of an otherwise complete statutory scheme,115 nor perform a new administrative function necessitated by novel regulatory legislation.
*263Whether or not federal criminal penalties should be redrafted116 it is of dubious constitutional propriety to delegate so crucial a legislative function to a non-representative body with no standards other than a direction that the results “not depreciate the seriousness of the offense” and “be consistent with the public welfare.” Accordingly, we decline to read such a delegation into the PCRA.117
This is not, of course, to say that no guidelines of any sort may issue under the PCRA, for some guidelines were clearly intended. Indeed such guidelines can constitute a laudable means of avoiding arbitrary decisionmaking. Issues of separation of powers involve subtle inquiries, sensitive to the precise dimensions of the power being exercised. Thus, if the guidelines function as the Commission contends, guiding but not controlling discretion, a different issue would be presented. They would then partake less of the character of fixed penalties than if Geraghty is correct, and would perform functions less like those of legislation. And guidelines which use the actual sentence imposed as a starting point would be akin to the exercise of sentencing discretion which the Congress may validly delegate to the courts. Such questions, however, may be disposed of only on a full record setting forth the actual functioning of the guidelines.
B. Ex Post Facto Effect of the Guidelines
At the time he was sentenced in the late summer of 1973, Geraghty was eligible for parole as soon as he was incarcerated.118 The grant or denial of parole, according to the then-current statute, was to result from a determination by the Board whether: (1) there is reasonable probability that the prisoner will live and remain at liberty without violating the laws; and (2) such release in the opinion of the Board is not incompatible with the welfare of society.119
At the time of his first parole hearing, guidelines adopted by the Commission subsequent to Geraghty’s sentence established a “customary time to be served before release” of 26 to 36 months for prisoners who had committed crimes as “severe” as the one for which Geraghty was convicted. And, claims Geraghty, less than one prisoner in ten is paroled before his “customary release date.”120 Consequently, Geraghty challenges the application of the parole guidelines to him, and others similarly situated, as constitutionally forbidden ex post facto laws.
1. Outlines of Ex Post Facto Prohibition
By the time of the American Revolution, Blackstone had classed ex post facto laws with those of Caligula: “all punishment [for violating them] must be cruel and unjust.” 121 In response to similar sentiments, as well as recent experience with oppressive ex post facto laws,122 the framers of the *264Constitution considered the necessity of guarding against such excesses strongly enough to include within the Constitution two prohibitions against ex post facto laws.123
These “bulwarks in favor of the personal security of the subject” were early read to be exclusively limitations on the exercise of the power to punish crimes.124 According to the celebrated case of Calder v. Bull, the interdiction included laws “aggravating a crime or mak[ing] it greater than it was, when committed” and laws “changing] the punishment, and inflictpng] a greater punishment than the law annexed to the crime, when committed.”125
But within the criminal sphere, the Supreme Court opted for a broad reading of the proscriptions: the definition of Calder was not exclusive, but merely indicative. Any law
passed after the commission of an offense which ... ‘in relation to that offense, or its consequences, alters the situation of a party to his disadvantage’ is an ex post facto law.126
Such alteration has been held to include not only liability for greater penalties, but aggravation of the conditions of the punishment imposed.127 The test is not whether the punishment actually received is within the outer limits of the law at the time the crime was committed, but whether “the later standard of punishment is more onerous than the earlier.”128
Thus, in Lindsey v, Washington, the Supreme Court held that a statute which altered a criminal penalty from a 1 to 15-year sentence to a 15-year indeterminate sentence with possibility of parole operated as an ex post facto law when applied to a crime which had taken place before the amendment. The Court said:
It is true that the petitioner might have been sentenced to 15 years under the old statute. But the ex post facto clause looks to the standard of punishment prescribed by the statute, rather than to the actual punishment imposed. The Constitution forbids the application of any new punitive measure to a crime already consummated to the detriment or material disadvantage of a wrongdoer, [citations omitted] It is for this reason that an increase in the possible penalty is ex post facto [citations omitted] ... regardless of the length of the penalty actually imposed.
We need not inquire whether this is technically an increase in the punishment annexed to the crime, See Calder v. Bull, 3 Dallas at 390. It is plainly to the substantial disadvantage of the petitioners to be deprived of all opportunity to receive a sentence which would give them freedom from custody and control prior to the expiration of their 15 year term.129
*2652. Ex Post Facto and Parole
The plaintiff in this ease had no more guarantee, of receiving parole before his “customary release date” than did the petitioners in Lindsey of receiving less than a maximum sentence. Like the adoption of a mandatory 15 year sentence in Lindsey, however, the promulgation of a 26 month “customary release date” deprived Ger-aghty of the possibility of a substantially more lenient punishment. The similarity between Geraghty’s position and that of the petitioners in Lindsey elicits concern whether the guidelines are at odds with the ex post facto prohibition. Our doubts are heightened by the two Supreme Court determinations which have touched upon the subject of parole and the ex post facto clause.
In Warden v. Marrero,130 the Court held that for purposes of a repeal of a statute barring parole for drug offenders, parole eligibility is effectively determined at the time of sentencing, and that therefore eligibility for parole is part of the “punishment” within the meaning of the statute.131 The Court, however, gave two additional reasons for treating parole eligibility as part of the punishment:
First, only an unusual prisoner could be expected to think that he was not suffering a penalty when he was denied eligibility for parole, [citations omitted] For the eonvicted prisoner, parole — even with its legal constraints — is a long step toward regaining freedom. .
Second, a repealer of parole eligibility previously available to imprisoned offenders would clearly present the serious question under the ex post facto clause of Art. I § 9 cl. 3, of the Constitution, of whether it imposed ‘greater or more severe punishment than was prescribed by law at the time of the ... offense.’ ” [citations omitted]132
Similarly, in Greenfield v. Scafati,133 a three-judge court struck down a statute depriving parole violators of accumulated good time upon their return to prison, as applied to a prisoner who had been sentenced before the law went into effect. Since the possible loss of good time for parole violation was in effect a potential lengthening of the sentence, the Court held that imposition of such a penalty with respect to the petitioner was an impermissible ex post facto law. In response to the argument that the plaintiff had, by his acceptance of parole, acceded to the condition, the Court declared:
[W]e see no distinction between depriving a prisoner of the right to earn good conduct deductions and the right to qualify for, and hence earn, parole. Each, to quote In re Medley, supra [134 U.S. 160, 10 S.Ct. 384, 33 L.Ed. 835], “materially alters the situation of the accused to his disadvantage.”134
Thus, the Court held, imposing the risk of loss of good time as a condition for parole violated the ex post facto prohibition. The Supreme Court affirmed without opinion.135
In light of Marrero and Scafati, along with the ex post facto decisions canvassed earlier, we conclude that were a statute to deprive already incarcerated or sentenced prisoners of the possibility of parole for the first 26 months of their sentence, when such was not the situation at the time of the sentence, that statute would be unconstitutional.136
*266The two courts of appeals which have considered the retroactive effect of the parole guidelines have reached conclusions similar to the ones thus far articulated. In Shepard v. Taylor,137 the court considered the applicability of the parole guidelines to prisoners sentenced under the Youth Corrections Act. Judge Kaufman held that “official post-sentence action that delays eligibility for supervised release runs afoul of the ex post facto proscription.” 138 Since the parole guidelines give substantial weight to offense severity, a factor which had been excluded from a decision to release under the Youth Corrections Act as it read at the time of sentencing, the Court determined that application of the guidelines to prisoners incarcerated under the original Youth Corrections Act would be unconstitutional. Likewise, in Ruip v. United States,139 the court held that “In any practical analysis, parole consideration is a part of the law annexed to the crime.”140
3. Ex Post Facto and the Guidelines
Both Shepard and Ruip, however, stated that the guidelines could be applied retroactively to adult offenders without trenching upon the ex post facto prohibition. This was so, the courts said, since the guidelines “merely clarify the exercise of administrative discretion,”141 and “the commission remains free to make parole decisions outside of these guidelines.”142 We respectfully differ with this conclusion.
From our analysis above, it is clear that the legislature could not retroactively, without offending the ex post facto clause of the Constitution, substantially decrease parole eligibility by legislation. We take it that a similar prohibition applies to an increase in punishment brought about by rule-making, the administrative equivalent of legislation.143 The legislature cannot, by delegation, escape constitutional limitations on its power. And to maintain that the present parole guidelines enact no substantial limitation on parole eligibility may well be to misinterpret both the legal and practical effects of the guidelines.
The legal effects of the guidelines prior to the PCRA were examined in Pickus v. U.S. Board of Parole,144 There, in holding the promulgation of the guidelines to be subject to the rule-making provisions of the Administrative Procedure Act, the court explicitly rejected the contention that these guidelines were mere general policy statements. Rather, in comparison with earlier rules which were themselves “calculated to have a substantial effect on ultimate parole *267decisions,” the guidelines “had' an even greater impact on an inmate’s chances for parole.”
The rules defining parole selection criteria are substantive agency action, for they define a fairly tight framework to circumscribe the Board’s statutorily broad power. . . [T]hey are self-imposed controls over the manner and circumstances in which the agency will exercise its plenary power. They have the force of law.145
As previously noted in the legislative history of the PCRA, the joint conference committee explicitly approved “the principles established in Pickus” with regard to the nature of the guidelines.146 Moreover, the PCRA itself empowers the Commission to “grant or deny release on parole notwithstanding the guidelines . . if it determines there is good cause for so doing.” 147 “Good cause,” however, according to the legislative history “for purposes of this section, means substantial reason.”148 It appears, therefore, that the guidelines are designed to be not mere hortatory “clarifications” of policy, but rules which are to be followed except for “substantial reason” to the contrary. The freedom of the Commission to grant parole prior to the customary release date is constrained, and the "situation” of prisoners otherwise eligible for parole is “altered to their disadvantage.”
The ex post facto clause, however, reaches beyond mere formal alterations in criminal law.149 Thus, if in practice the parole authorities found good cause to deviate from the guidelines in 60% of the cases, for example, it might be argued that their discretion is, in fact, unfettered. However, the material which we have before us points in the opposite direction. The Parole Commission’s answer to the complaint in this case admits that in 1975 prisoners were granted parole prior to their “customary release dates” in only 8.7% of the cases.150 Another decision of this Court took judicial notice of estimates of compliance with the guidelines ranging from 88% to 94%.151 It thus appears that the “channel for discretion” provided by the guidelines is in actuality an unyielding conduit.
Nonetheless, since the Commission intimates elsewhere in its answer that, in fact, it engages in individualized consideration of prisoners similar to that which it undertook before the guidelines went into effect, there appears to be a controverted issue of material fact, and on such issue, the case is inappropriate for summary judgment.
IV. CONCLUSION
(1) The federal courts have jurisdiction over the complaint filed by Geraghty since the suit was properly brought as an action for declaratory judgment under the APA and the PCRA.
*268(2) If the class certification was improperly denied, a live legal controversy still exists despite Geraghty’s release, and the action is not moot.
(3) It was error for the trial court to fail to consider the possibility of creating subclasses out of the over-inclusive and heterogeneous classes specified by the plaintiff, and therefore the case must be remanded to the trial court to evaluate the possibility of creating subclasses.152
(4) The substantive claims set forth in the complaint were not properly subject to dismissal by summary judgment. Plaintiff’s contentions are supported by allegations sufficient to raise genuine issues of
fact, and, if the averments are accurate, the parole guidelines as administered may well be inconsistent with the PCRA, and the parole guidelines as applied to certain prisoners may violate the ex post facto prohibition.
(5) If on remand the trial court determines that a proper class action should be maintained, factual support for the plaintiff’s allegations regarding the operation of the parole system must be fully developed, and the Commission should have ample opportunity to establish the authenticity of its version of the operation of the guidelines.
The ease will be reversed and remanded for action in accordance with this opinion.
[[Image here]]
*269[[Image here]]
*270[[Image here]]
*271[[Image here]]
. The Act of June 25, 1910, ch. 387 § 1, 36 Stat. 819, was the first legislation that established authority to grant parole to federal prisoners.
. 38 Fed.Reg. 26652-57 (Sept. 24, 1973). For an extensive analysis of the guidelines and associated issues of policy and law, see Note, Parole Release Decisionmaking and the Sentencing Process. 84 Yale L.J. 810 (1975) [hereinafter cited as Yale Note].
. 18 U.S.C. §§ 4201 — 4218 (1976).
. 18 U.S.C. § 4203(a) (1976).
. 18 U.S.C. § 4206 (1976).
. The current guidelines, codified at 28 C.F.R. § 2.20, are reproduced as Appendix I, infra. Previous cases in this Circuit concerning the guidelines have dealt primarily with the availability of 28 U.S.C. § 2255 as a vehicle for trial judges to resentence prisoners where the expectations of the sentencing judge have been frustrated by the applications of the guidelines. Addonizio v. United States, Nos. 77-1542, 77-1621, 77-2373, 573 F.2d 147 (3d Cir. 1978); United States v. Somers, 552 F.2d 108 (3d Cir. 1977); United States v. Salerno, 538 F.2d 1005 (3d Cir.) aff'd per curiam sur petition for rehearing 542 F.2d 628 (3d Cir. 1976); cf. United States v. Solly, 559 F.2d 230 (3d Cir. 1977) (reversing denial of Rule 35 motion where parole guidelines operated to frustrate sentencing judge’s expectation). The availability of § 2255 has been a subject of some disagreement among the circuits. Compare e. g. United States v. Kent, 563 F.2d 239 (5th Cir. 1977); United States v. McBride, 560 F.2d 7 (1st Cir. 1977); Andrino v. United States Board of Parole, 550 F.2d 519 (9th Cir. 1977) with e. g. Kortness v. United States, 514 F.2d 167 (8th Cir. 1975); United States v. Slutsky, 514 F.2d 1222 (2d Cir. 1975).
Recently, in United States v. Musto, No. 77-1239, 571 F.2d 136 (1978), this Court held that § 2255 was unavailable as a jurisdictional base for an attempt by one already confined to alter a sentence imposed by a judge who had knowledge of the parole guidelines. But Musto would not appear to alter the holding in Zanni-no v. Arnold, 531 F.2d 687, 690 (3d Cir. 1976), that judicial review of a parole board decision is available under 28 U.S.C. § 2241 “to insure that the Board has followed criteria appropriate, rational and consistent with the statute and that its decision is not- arbitrary and capricious nor based on impermissible considerations.” See id. at 689 n.5.
. Geraghty’s conviction was affirmed in United States v. Braasch, 505 F.2d 139 (7th Cir. 1974), cert. denied sub nom. Geraghty v. United States, 421 U.S. 910, 95 S.Ct. 1561, 43 L.Ed.2d 775 (1975). His sentence was subsequently reduced by the trial court from 4 years to 30 months, under Rule 35, F.R.Crim.P. The reduction was based on a finding that the guidelines would cause Geraghty to be imprisoned substantially longer than the district court intended. United States v. Braasch, No. 72 C.R. 979 (N.D.Ill. Oct. 1, 1975), mandamus denied 542 F.2d 442 (7th Cir. 1976).
. A verbatim reproduction of this statement was upheld by a panel of this Circuit as a sufficient articulation of reasons to satisfy the demands of due process. Hill v. Atty. Gen., 550 F.2d 901 (1977); accord Garcia v. United States, 557 F.2d 100 (7th Cir. 1977).
. Although the action was originally filed in the District Court for the District of Columbia, the matter was transferred to the Middle District of Pennsylvania, where Geraghty was then confined. Geraghty v. United States Parole Commission, Civ. No. 76-1729 (D.D.C.1976).
. 411 U.S. 475, 93 S.Ct. 1827, 36 L.Ed.2d 439 (1973).
. 28 U.S.C. § 2254(b) (1970).
. 411 U.S. at 490-92, 93 S.Ct. 1827.
. 411 U.S. at 489-90, 93 S.Ct. at 1836.
. 18 U.S.C. § 4218 became effective May 15, 1976. The suit in this case was filed in September of 1976.
. House Conf.Rep.No.94-838, 94th Cong. 2nd Sess. 36 reprinted 1976 U.S.Code, Cong. & Admin.News pp. 351, 368. [hereinafter cited as Conference Report]
. 165 U.S.App.D.C. 284, 507 F.2d 1107 (1974).
. This expression of intent by Congress renders inapplicable the Supreme Court’s decision in Califano v. Sanders, 430 U.S. 99, 97 S.Ct. 980, 51 L.Ed.2d 192 (1977). Califano held that as a matter of statutory construction, the APA does not confer jurisdiction on federal courts, absent other statutory authorization.
. Bijeol v. Benson, 513 F.2d 965, 967 (7th Cir. 1975) and Billiteri v. U.S. Parole Bd., 541 F.2d 938, 947-48 (2d Cir. 1976), may appear to be at odds with our decision. The Court in Bijeol, however, did not have before it the provisions of 18 U.S.C. § 4218. And in Billiteri the plaintiff requested neither declaratory nor prospec-five injunctive relief, and did not challenge the propriety of the guidelines.
This Court’s holding in Marerro v. Warden, 483 F.2d 656, 659-60 (3d Cir. 1973), rev’d on other grounds 417 U.S. 653, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974), that habeas corpus may be used to attack a denial of parole does not undermine our result, for we in no way intimated in Marerro that habeas corpus was the sole route available. Cf. Preiser v. Rodriguez, 411 U.S. 475, 498-99, 93 S.Ct. 1827, 36 L.Ed.2d 439 (1973) (habeas and § 1983 are alternative remedies with respect to prison conditions).
. See Board of School Commissioners of Indianapolis v. Jacobs, 420 U.S. 128, 95 S.Ct. 848, 43 L.Ed.2d 74 (1975); Weinstein v. Bradford, 423 U.S. 147, 96 S.Ct. 347, 46 L.Ed.2d 350 (1975) ; Pasadena Bd. of Educ. v. Spangler, 427 U.S. 424, 430, 96 S.Ct. 2697, 49 L.Ed.2d 599 (1976) .
. See, e. g. Super Tire Engineering Co. v. McCorkle, 416 U.S. 115, 94 S.Ct. 1694, 40 L.Ed.2d 1 (1974); United States v. New York Telephone Co., 434 U.S. 159, 165 n.6, 98 S.Ct. 364, 54 L.Ed.2d 376 (1977); United States v. Frumento, 552 F.2d 534 (3d Cir. 1977) (en banc).
. See, e. g. Gerstein v. Pugh, 420 U.S. 103, 110 n.ll, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975); Williams v. Wohlgemuth, 540 F.2d 163, 167 (3d Cir. 1976).
The plaintiff also presses the application of Eliezer Becher to intervene as a named plaintiff. Becher, like Geraghty, has been an inmate of Lewisburg who has been denied parole. Unlike Geraghty, however, Becher had not been released at the time his motion to intervene was argued before this Court. Becher’s motion was originally presented after the appeal to this Court had been docketed, and the district court divested of jurisdiction. His request to intervene was therefore denied. Since we conclude that the case must be remanded to the district court for other determinations, it is also appropriate that Becher’s motion to intervene be remanded for a determination as to the reasons for his failure to intervene earlier, and an examination of the potential prejudice, if any, which might result from such intervention. See Pennsylvania v. Rizzo, 530 F.2d 501 (3d Cir.), cert. denied 426 U.S. 921, 96 S.Ct. 2628, 49 L.Ed.2d 375 (1976). Cf. United Airlines v. McDonald, 432 U.S. 385, 97 S.Ct. 2464, 53 L.Ed.2d 423 (1977).
. The development of the principles governing mootness and their interaction with the emergence of class actions has been the subject of a considerable number of recent scholarly discussions. See H. Newberg, Newberg on Class Actions §§ 1085-1092 (1977); Kane, Standing, Mootness and Federal Rule 24 — Balancing Perspectives, 26 Buffalo L.Rev. 83 (1976); Developments in the Law — Class Actions, 89 Harv.L. Rev. 1163-71 (1976); Comment, Continuation and Representation of Class Actions Following Dismissal of the Class Representative, 1974 Duke L.J. 573; The Mootness Doctrine in the Supreme Court, 88 Harv.L.Rev. 373 (1974); Note, Mootness on Appeal in the Supreme Court, 83 Harv.L.Rev. 1672 (1970); Note, A Search for Principles of Mootness in the Federal Courts, Part One, the Continuing Impact Doctrine, 54 Texas L.Rev. 1289 (1976); Part Two, Class Actions id. 1310.
. See Note, The Mootness Doctrine, supra note 22, 373 and 374-75 nn.9-11 (1974); Note, Mootness on Appeal, supra note 22, at 1672, 1673-74 n.12 (and authorities cited therein).
. U.S.Const. Art. III, § 2. The first mootness holding explicitly couched in terms of Art. III jurisdictional limitations was Sibron v. New York, 392 U.S. 40, 57, 88 S.Ct. 1889, 20 L.Ed.2d 917 (1968), although the Court had referred to Art. III limitations in Liner v. Jafco, 375 U.S. 301, 306 n.3, 84 S.Ct. 391, 11 L.Ed.2d 347 (1964) and Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 240, 57 S.Ct. 461, 81 L.Ed. 617 (1937). See Note, Mootness Doctrine, supra note 22, at 375, Note, Mootness on Appeal, supra note 22 at 1673-74; cf. G. Gunther, Constitutional Law 1578-80 (9th Ed. 1975) (evidencing skepticism about recent assertion of constitutional origins of mootness doctrine).
. Franks v. Bowman Transportation Co., 424 U.S. 747, 755-57 and n.8, 96 S.Ct. 1251, 47 L.Ed.2d 444 (1976) (for a unanimous court on this point); Kremens v. Bartley, 431 U.S. 119, 127-132, 97 S.Ct. 1709, 52 L.Ed.2d 184 (1977). See Kane, supra note 22, at 84-88, 94.
. Flast v. Cohen, 392 U.S. 83, 94-95, 88 S.Ct. 1942, 1950, 20 L.Ed.2d 947 (1968), quoted in Franks v. Bowman Transportation Co., 424 U.S. 747, 755, 96 S.Ct. 1251, 47 L.Ed.2d 444 (1976). See e. g. Note, Mootness Doctrine, supra note 22 at 376-377.
. The judgment of a federal court must resolve “a real and substantial controversy admitting of specific relief through a decree of a conclusive character, as distinguished from an opinion advising what the law would be upon a hypothetical state of facts.” Preiser v. Newkirk, 422 U.S. 395, 401, 95 S.Ct. 2330, 2334, 45 L.Ed.2d 272 (1975), quoting North Carolina v. Rice, 404 U.S. 244, 246, 92 S.Ct. 402, 30 L.Ed.2d 413 (1971), quoting Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 241, 57 S.Ct. 461, 81 L.Ed. 617 (1937).
. See, e. g„ Warth v. Seldin, 422 U.S. 490, 498-501, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975).
. See Kane, supra note 22 at 84-85, 94-96 (two part test exists: actual injury and controversy and discretionary determination), Note, Mootness Doctrine, supra note 22 at 389, 394-95 (discretion should look to impact of a decision on the merits on the necessity for future litigation. The amenability of a fact pattern to definitive resolution would imply that future litigation might be minimized by taking jurisdiction).
. See Note, Developments in the Law — Class Actions, 89 Harv.L.Rev. 1318, 1364-66, (1977) (Supreme Court has moved from the focus on continuing controversy with named plaintiff to an inquiry into the status of legal issues in dispute between the class and the class adversary); Note, Mootness Doctrine, supra note 22 at 388-395 (non-mootness may be predicated upon existence of injury to class).
. 552 F.2d 534 (3d Cir. 1977) (en banc).
. E. g. Pacific Terminal Co. v. ICC, 219 U.S. 498, 31 S.Ct. 279, 55 L.Ed. 310 (1911) (regulatory order effective for two years); Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973) (gestation period as limit to impact of abortion law); Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532 (1976) (one year residency requirement for divorces); Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972) (one year residency period for eligibility to vote); Nebraska Press Assn. v. Stuart, 427 U.S. 539, 96 S.Ct. 2791, 49 L.Ed.2d 683 (1976) (“gag order” limited to duration of trial); Super Tire Engineering Co. v. McCorkle, 416 U.S. 115, 125-27, 94 S.Ct. 1694, 40 L.Ed.2d 1 (1974) (welfare benefits to striking workers); United States v. New York Telephone Co., 434 U.S. 159, 165 n. 6, 98 S.Ct. 364, 54 L.Ed.2d 376 n. 6 (1977) (“pen register” order of limited duration). Gerstein v. Pugh, 420 U.S. 103, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975) (pretrial detention). See Note, Mootness Doctrine, supra note 22 at 383-388, Singleton v. Wulff, 428 U.S. 106, 117, 96 S.Ct. 2868, 49 L.Ed.2d 826 (1976) (dictum).
Cf. Preiser v. Newkirk, 422 U.S. 395, 95 S.Ct. 2330, 45 L.Ed.2d 272 (1975) (challenge to prison disciplinary procedure moot where punishment of named plaintiff under procedure had ceased, no class action had been alleged, and plaintiff could have no “reasonable expectation of repetition” of the procedure in question).
. Compare e. g. United States v. New York Telephone Co., 434 U.S. 159, 98 S.Ct. 364, 54 L.Ed.2d 376 (1977); Nebraska Press Assn. v. Stuart, 427 U.S. 539, 96 S.Ct. 2791, 49 L.Ed.2d 683 (1976); Super Tire Engineering Co. v. McCorkle, 416 U.S. 115, 94 S.Ct. 1694, 40 L.Ed.2d 1 (1974) with Warth v. Seldin, 422 U.S. 490, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975); Simon v. Eastern Ky. Welfare Rights Organization, 426 U.S. 26, 96 S.Ct. 1917, 48 L.Ed.2d 450 (1976).
. See Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973) (other pregnant women denied abortion, even after plaintiffs term of pregnancy); Sosna v. Iowa, 419 U.S. 393, 399-403, 95 S.Ct. 553, 42 L.Ed.2d 532 (1975) (1 year residency requirement for divorce, applicable to other potential divorcees); Dunn v. Blum-stein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972) (1 year voter residency requirement applicable to others who move into the voting district).
. Gerstein v. Pugh, 420 U.S. 103, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975); Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532 (1975); Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973); Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972).
. 418 U.S. 24, 94 S.Ct. 2655, 41 L.Ed.2d 551 (1974).
. 418 U.S. at 35, 94 S.Ct. 2655.
. See Warth v. Seldin, 422 U.S. 490, 514-517, 95 S.Ct. 2197, 45 L.Ed.2d 343 (1975).
. 424 U.S. 747, 96 S.Ct. 1251, 47 L.Ed.2d 444 (1976).
. 424 U.S. at 756 n. 8, 96 S.Ct. 1251. See 424 U.S. at 781, 96 S.Ct. 1251 (opinion of Powell, J. joined by Rehnquist, J. concurring on this point).
. 424 U.S. at 756-57, 96 S.Ct. 1251. In contrast, in E. Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 97 S.Ct. 1891, 52 L.Ed.2d 453 (1977), where the plaintiff had never been a part of the class which he sought to represent, it was held that the class was improperly certified on appeal. Cf. Kremens v. Bartley, 431 U.S. 119, 97 S.Ct. 1709, 52 L.Ed.2d 184 (1977), where the Supreme Court held that changes in law so fragmented the interest of the plaintiff class that it could no longer clearly identify a common interest.
. 420 U.S. 128, 95 S.Ct. 848, 43 L.Ed.2d 74 (1975).
. The statements in Sosna, Franks, Gerstein, and East Texas that justiciability remains “given a properly certified class” may simply point to certifiability, not actual certification, as the crucial question. See Satterwhite v. City of Greenville, 557 F.2d 414, 417-421 (5th Cir. 1977) rehearing en banc granted; Frost v. We-inberger, 515 F.2d 57, 64 (2d Cir. 1975) (Friendly, J.) (Sosna’s insistence on continuing existence of individual claim until certification is “drained” by “relation back” doctrine). Similarly, Weinstein v. Bradford, 423 U.S. 147, 96 S.Ct. 347, 46 L.Ed.2d 350 (1975), which dismissed a parole challenge as moot, is inappo-site here since the Weinstein plaintiffs did not appeal the denial of class certification. Cf. Scott v. Kentucky Parole Board, 429 U.S. 60, 97 S.Ct. 342, 50 L.Ed.2d 218 (1976) (individual action, no attempt to certify class).
. 420 U.S. 103, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975).
. The stated rationale, that the later certification “related back” to the time that the plaintiffs had claims is at best a legal fiction. See Gardner v. Westinghouse, (Seitz, J., concurring) No. 76-1410, 559 F.2d 209, at 216-217 (3d Cir. 1977), aff’d- U.S. -, 98 S.Ct. 2451, 57 L.Ed.2d 364 (1978), Frost v. Weinberger, 515 F.2d 57, 64 (2d Cir. 1975).
The Gerstein decision had been presaged by Kelly v. Wyman, 294 F.Supp. 887, 890 (S.D.N. *250Y.1968) aff’d sub nom. Goldberg v. Kelly, 397 U.S. 254, 256 n. 2, 90 S.Ct. 1011, 25 L.Ed.2d 287 (1970). There the district court refused to dismiss a challenge to welfare regulations before class certification despite an allegation that most if not all of the named plaintiffs had been reinstated to welfare, on the ground that “plaintiffs claim to represent a class.” The Supreme Court affirmed the class relief, noting that some of the then-named plaintiffs were still embroiled in disputes with the welfare department. It was not clear, however, whether those plaintiffs had been named at the time of the original decision or had subsequently intervened. See Goosby v. Osser, 409 U.S. 512, 93 S.Ct. 854, 35 L.Ed.2d 36 (1973) (adjudicating class action of prisoners confined to pretrial detention, despite the fact that apparently there was no indication whether they had been released prior to class certification); Conover v. Montemuro, 477 F.2d 1073 (3d Cir. 1973) (en banc) (allowing class action on behalf of juvenile detainees, despite the fact that named detainees had been released after short time).
Since Gerstein a number of courts have allowed certification on a “relating back” theory. E. g. Williams v. Wohlgemuth, 540 F.2d 163 (3d Cir. 1976); Basel v. Knebel, 179 U.S.App.D.C. 209, 211, 551 F.2d 395, 397 (1977); Zurak v. Regan, 550 F.2d 86 (2d Cir. 1977), cert. denied, 433 U.S. 914, 97 S.Ct. 2988, 53 L.Ed.2d 1101.
. 425 U.S. 308, 311 n. 1, 96 S.Ct. 1551, 47 L.Ed.2d 810 (1976).
. 432 U.S. 385, 97 S.Ct. 2464, 53 L.Ed.2d 423 (1977). Cf. Haas v. Pittsburgh National Bank, 526 F.2d 1083 (3d Cir. 1975) (allowing substitute plaintiff to intervene for named plaintiff who never had standing, but allowing filing of original class action to toll statute of limitations).
. As other courts have noted, the failure to certify the class in Jacobs was not assigned as error. Napier v. Gertrude, 542 F.2d 825, 827 (10th Cir. 1976) cert. denied 429 U.S. 1049, 97 S.Ct. 759, 50 L.Ed.2d 765 (1977); Gardner v. Westinghouse, No. 76-1410, 559 F.2d 209, at 214 n. 5 (3d Cir. 1977) (Seitz, J„ concurring) aff’d- U.S.---, 98 S.Ct. 2451, 57 L.Ed.2d 364(1978).
. We acknowledge that the courts of appeals are divided on the question of whether under the recent Supreme Court decisions, the denial of class action status is appealable by a named plaintiff whose claim has become moot. Banks v. Multi Family Management, 554 F.2d 127 (4th Cir. 1977) (Refusal to certify was not appeala-ble since development of mootness was not “inherent in the nature of the claims,” and the named defendant had already agreed to relief sought); Satterwhite v. City of Greenville, 557 F.2d 414 (5th Cir. 1977) rehearing granted en banc (refusal to certify class is appealable if improper if at the time that the class should have been certified, and the named plaintiff still had a live claim at that time); Kuahulu v. Employers Insurance of Wausau, 557 F.2d 1334 (9th Cir. 1977) (survival of action is governed by “idiosyncracies of each case;” action in question did not survive, but the Court suggests a number of situations in which it might); Winokur v. Bell Federal Savings and Loan, 560 F.2d 271 (7th Cir. 1977) (where damages claimed were tendered to individuals, nonplau-sible claim of repetition, no review allowed of failure to certify class. The court enunciated dictum that issue of class certification cannot survive mooting of individual claim); Napier v. Gertrude, 542 F.2d 825 (10th Cir. 1976) (review available where failure to certify is “correctable on appeal,” correctability is a function of particular situations, mainly available in situations otherwise evading review).
. 18 U.S.C. § 4205 (1976).
. See Zurak v. Regan, 550 F.2d 86 (2d Cir. 1977) (prisoners whose complaints on parole procedure would often — but not always— evade review were allowed to “relate back” certification to a time before the named plaintiff’s release). Cf. United States v. Frumento, 552 F.2d 534 (3d Cir. 1977) (en banc) (compulsion to testify before grand jury).
. The government’s brief in this action (p. 12) suggests that only prisoners with short sentences have an interest in attacking the guidelines’ failure to take account of sentence length.
. No. 76-1410, 559 F.2d 209 (3d Cir. 1977) (Seitz, C. J. concurring), aff’d-U.S.-, 98 S.Ct. 2451, 57 L.Ed.2d 364 (1978).
. Cf. Gerstein v. Pugh, 420 U.S. 103, 110, 111 n. 11, 95 S.Ct. 854, 43 L.Ed.2d 54 (1975) (public defender had continuous relationship with class of pretrial detainees).
. Cf. Kremens v. Bartley, 431 U.S. 119, 97 S.Ct. 1709, 52 L.Ed.2d 184 (1977).
. See Note, Mootness Doctrine, supra note 22 at 394. Cf. De Funis v. Odegaard, 416 U.S. 312, 94 S.Ct. 1704, 40 L.Ed.2d 164 (1975).
. A. 46.
. 18 U.S.C. § 4208(a)(2) (1970).
. Judge Herman also held that the district court’s habeas corpus jurisdiction did not extend to members of the class located outside of the Middle District of Pennsylvania. Since we hold that jurisdiction lies to consider this action as one for declaratory judgment, we need not reach this question.
. Complaint, 5(a)(A2).
. Section 23(c)(4) reads: “When appropriate (A) an action may be brought or maintained as a class action with respect to particular issues, or (B) a class may be divided into subclasses and each subclass treated as a class, and the provisions of this rule shall then be construed and applied accordingly.”
. See, e. g., Brown v. United States, 508 F.2d 618, 627 (3d Cir. 1975) (affirming limitation of class in part) cert. denied 422 U.S. 1027, 95 S.Ct. 2621, 45 L.Ed.2d 684 (1975); Swarb v. Lennox, 314 F.Supp. 1091, 1098-99 (E.D.Pa. 1970) (three-judge court, limiting class sua sponte) aff’d in part, 405 U.S. 191, 92 S.Ct. 767, 31 L.Ed.2d 138 (1972); C. Wright & A. Miller, Federal Practice & Procedure § 1759 pp. 575-76 (“ . . . if the plaintiffs definition of the class is found to be unacceptable, the court may construe the complaint or redefine the class to bring it within the scope of Rule 23”); Id. at § 1790; Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 179-186, 94 S.Ct. 2140, 40 L.Ed.2d 732 (Douglas. J., concurring and dissenting); AAMCO Automatic Transmissions, Inc. v. Tayloe, 407 F.Supp. 430 (E.D.Pa. 1976) (redefining class). Note, Developments in the Law, supra note 22 at 1479-93 (discussing possibilities of sub-classing and class redefinition).
. 538 F.2d 991 (3d Cir. 1976).
. 538 F.2d at 996. See Bogosian v. Gulf Oil Corp., 561 F.2d 434, 453 (3d Cir. 1977) cert. denied 434 U.S. 1086, 98 S.Ct. 1280, 55 L.Ed.2d 791 (“even assuming that the court were correct in its conclusion that the lease claim is not appropriate for class determination, it nevertheless should have considered certification of the trademark claim under Rule 24(c)(4)(a)”); Wright & Miller, supra note 62 at § 1790, pp. 185-87. (“Rule 23(c)(4) imposes a duty on the court and gives it ample power . . . it is not bound by the plaintiffs complaint and should not dismiss the action simply because it misdefines the class or issues when the court can correct the situation under 23(c)(4)”). Newberg, Class Actions, supra note 22 at § 1120h (if conflicting interests can be protected by sub-classification, conflict should not preclude class action).
. Cf. Wetzel v. Liberty Mutual Ins. Co., 508 F.2d 239, 253 (3d Cir. 1975) (subclassification “is required where the class includes subclasses with divergent interests, or where certain representatives adequately represent only one group and other representatives represent another group.”) cert. denied 421 U.S. 1011, 95 S.Ct. 2415, 44 L.Ed.2d 679 (1975).
. In addition to determining the proper boundaries of the subclasses in this case, the district court must ascertain whether the other prerequisites of class certification have been met. It may also wish to seek amici curiae to represent divergent interests of subclasses.
. While the ex post facto contention was not pressed strongly on appeal, it formed a central element of the complaint and the district court’s opinion. Moreover, challenges on appeal to the “retroactive” effect of PCRA (plaintiff’s brief p. 57) seem to be rooted in the ex post facto prohibition,
Geraghty also alleges that the guidelines conflict with the Congressional intent underlying 18 U.S.C. § 4208(a)(2) (1970). If we determine that the guidelines are consistent with the Congressional intent in adopting the PCRA, this objection is substantially undercut, for the PCRA reenacts and recodifies the provisions of § 4208(a)(2) as 18 U.S.C. § 4205(b) (1976). Insofar as this is a change in the law, it can be analyzed under the discussion of the ex post facto objections.
. The “offenses” do not necessarily involve violations of different statutes. For example, income tax evasion (less than $10,000) is a different “offense” from income tax evasion ($10,000-50,000). Likewise, possession with intent to distribute “soft drugs” falls into categories ranging from low moderate to very high, depending on the amount of drugs possessed.
. These categories are identified by reference to “salient factor scores,” which attempt to predict the probability of recidivism.
. The statement of reasons which was given to Geraghty, for example, was duplicated in Garcia v. United States Bd. of Parole, 557 F.2d 100 (7th Cir. 1977); Hill v. Attorney General, 550 F.2d 901 (3d Cir. 1977) and Fronczak v. Warden, 553 F.2d 1219 (10th Cir. 1977). The Second Circuit has manifested disapproval of “mechanical” application of the guidelines by the Parole Board. United States v. Jackson, 550 F.2d 830, 832 (2d Cir. 1977); United States v. Cruz, 544 F.2d 1162, 1164-65 n. 6 (2d Cir. 1976); United States v. Torun, 537 F.2d 661, 664 (2d Cir. 1976).
. 28 C.F.R. § 2.19(b).
. 28 C.F.R. § 2.20(c).
. Government’s answer 37. A38.
. Defendant’s brief in opposition to petition for habeas corpus. (Filed Dec. 13, 1976) p. 19 n. 8. The Board had previously asserted that to take sentence length into account in the guidelines would impede the guidelines’ purpose of eliminating sentence disparity. 40 Fed. Reg. 41350 (1975). Cf. United States v. Solly, 559 F.2d 230, 233 (3d Cir. 1977) (neither sentence length nor trial judge’s assessment of culpability has “apparent effect” on release decision).
. However, since the Parole Commission has presented contradictory material, and since a large part of Geraghty’s proof is inferential, the case cannot be resolved by summary judgment in Geraghty’s favor on the basis of the present record.
. In making a determination under this chapter (relating to release on parole) the Commission shall consider, if available and relevant:
(1) reports and recommendations which the staff of the facility in which such prisoner is confined may make;
(2) official reports of the prisoner’s prior criminal record, including a report or record of earlier probation and parole experiences;
(3) presentence investigation reports;
(4) recommendations regarding the prisoner’s parole made at the time of sentencing by the sentencing judge; and
(5) reports of physical, mental, or psychiatric examination of the offender.
There shall also be taken into consideration such additional relevant information concerning the prisoner (including information submitted by the prisoner) as may be reasonably available.
18 U.S.C. § 4207 (1976).
. The primary basis for this assertion is a study of the North Carolina parole system which found sentence length to be a primary determinant of length of imprisonment. This is not persuasive evidence of the characteristics of the federal parole system. Nonetheless, one study found that the belief was widespread among federal judges prior to the guidelines that the Parole Board’s decisions were or should be based on judgments as to rehabilitation, rather than evaluations of severity. Yale Note, supra note 2 at 882-83 n. 361, 890 n. 388; see Addonizio v. United States, Nos. 77-1542, 77-1621, 77-2373, 573 F.2d 147 (3d Cir. 1978). And the Commission has presented no evidence which convinces us that the primacy accorded to its severity judgment is reflective of practice prior to the guidelines.
. The PCRA was reported from the House Judiciary Committee on May 13, 1975 as HR 5727, accompanied by H.Rept. 94-184. It was considered and passed by the House on May 21, 1975.
The Senate Judiciary Committee reported an amended version of H.R. 5727 on Sept. 11, 1975, accompanied by Sen.Rept. 94-369. The amended version was passed by the Senate on Sept. 16, 1975.
*257A House-Senate conference committee reported a compromise bill on Feb. 23, 1976, and Feb. 24, 1976, respectively. The amended bill passed the Senate on March 2, 1976, and the House on March 3, 1976. See 1976 U.S.Code Cong. & Admin.News p. 335.
. § 4205 (House version)
. Comments of Congressman Rodino, 121 Cong.Rec.H. 4510 (daily ed. May 21, 1975).
. See e. g. Comments of Congressman Kasten-meir id. at H. 4511; Comments of Congressman Railsback id. at H. 4511; Comments of Congressman Danielson id. at H. 4512; Comments of Congressman Badillo id. at H. 4516; Comments of Congressman Coghlin id. at H. 4516.
. Comments of Congressman Drinan at id. H. 4513 (“the committee clearly expressed its objection to a checklist or prototype denial statement”); Comments of Congressman Gude at id. at H. 4518 (Bill would produce “far greater scrutiny and understanding of each prisoner’s record, both before and during incarceration”). Rep. Gude was “greatly disturbed” by the apparently mechanical interaction of the “severity rating” and “salient factor score.”
. Remarks of Senator Burdick, 121 Cong. Rec.S. 15953 (daily ed. Sept. 16, 1975) (“the rules and regulations repeatedly cited in the legislation include a set of guidelines adopted by the Parole Board.”); Sen.Rept.No.94-369, 94th Cong. 1st Sess. 25 [hereinafter, Senate Report], 1976 U.S.Code Cong. & Adm.News p. 346 (describing operation of system contemplated by statute “in relation to the present guidelines system”).
Even the Senate version, however, recognized the danger of allowing too little discretion in the parole decision which “fails to take into account aggravating or mitigating circumstances of an offense, as well as the personal and family considerations which are the strongest predictors of whether or not the individual will return to a life of crime, or will be able to live a law-abiding life-style following release from incarceration.” (remarks of Senator Bur-dick, supra at S. 15954). Senator Burdick considered that the Senate Bill “incorporated these factors” id.
. § 4206(a). (Senate version)
. § 4206(b). (Senate version)
. Senate Report at p. 15, 1976 U.S.Code Cong. & Admin.News p. 337.
. Id. at 18-19 and 340.
. 122 Cong.Rec.H. 1499 (daily ed. March 3, 1976).
. Conference Report at 19-20, 1976 U.S.Code Cong. & Admin.News pp. 351-52.
. Id. at 19, 352 (emphasis supplied).
. We understand the conferees’ statement, that the statute “removes doubt as to the leg.dity of changes implemented by administrative reorganization, and makes the improvements permanent” id. at 20, 353, to refer to an endorsement of the use of guidelines in the context of a decentralized hearing examiner structure, not as an adoption of the substance of the guidelines, or their then-current use. In this regard, we respectfully disagree with the statements in Banks v. United States, 553 F.2d 37 (8th Cir. 1977) that the Committee approved the then-current guidelines and that a primary purpose of the PCRA was to reduce sentence disparity. Cf. Garcia v. United States Bd. of Parole, 557 F.2d 100 (7th Cir. 1977) (holding Parole Commission authorized by PCRA to take account of offense severity in guidelines, but declining to rule on whether guidelines are consistent with PCRA).
. Conference Report at 26, 1976 U.S.Code Cong. & Admin.News p. 359.
. Id. at 25, 358.
. Id. at 26, 358.
. Id.
. See Complaint; note 74 supra; Cf. United States v. Randle, 408 F.Supp. 5 (N D.Ill.1975). (“In effect, the Parole Board has substituted its guidelines for the judgment of the sentencing judge as to the appropriate sentence for a particular defendant.”)
. 40 Fed.Reg. 2331, August 1975.
. Conference Report at 25, 1976 U.S.Code Cong. & Admin.News p. 358. (emphasis added).
. See National Cable Television Assn., Inc. v. United States, 415 U.S. 336, 94 S.Ct. 1146, 39 L.Ed.2d 370 (1974).
. Woodson v. North Carolina, 428 U.S. 280, 304, 96 S.Ct. 2978, 2991, 49 L.Ed.2d 944 (1976) (plurality opinion of Justices Stevens, Stewart and Powell) (quoting Pennsylvania ex rel. Sullivan v. Ashe, 302 U.S. 51, 55, 58 S.Ct. 59, 82 L.Ed. 43 (1937) as characterizing “prevailing practice.”). See also Gregg v. Georgia, 428 U.S. 153, 189, 96 S.Ct. 2909, 49 L.Ed.2d 859 (1976) (plurality opinion of Justices Stevens, Stewart and Powell).
. Williams v. New York, 337 U.S. 241, 247, 69 S.Ct. 1079, 1083, 93 L.Ed. 1337 (1949); see Chaffin v. Stynchcombe, 412 U.S. 17, 21-22, 93 S.Ct. 1977, 36 L.Ed.2d 714 (1973).
. Dorszynski v. United States, 418 U.S. 424, 94 S.Ct. 3042, 41 L.Ed.2d 855 (1974); Gore v. United States, 357 U.S. 386, 393, 78 S.Ct. 1280, 2 L.Ed.2d 1405 (1958); United States v. Bazza-no, 570 F.2d 1120, at 1123-1124 (3d Cir. 1977) (Adams, J. concurring).
. See e. g. United States v. Negron, 548 F.2d 1085, 1087 (2d Cir. 1977), cert. denied, 433 U.S. 912, 97 S.Ct. 2981, 53 L.Ed.2d 1096; United States v. Foss, 501 F.2d 522 (1st Cir. 1974); United States v. Schwarz, 500 F.2d 1350 (2d Cir. 1974); Woolsey v. United States, 478 F.2d 139 (8th Cir. 1973) (en banc); United States v. Hartford, 489 F.2d 652 (5th Cir. 1974); United States v. Charles, 460 F.2d 1093 (6th Cir. 1972); United States v. Baker, 487 F.2d 360 (2d Cir. 1973); United States v. Daniels, 446 F.2d 967 (6th Cir. 1971); United States v. McCoy, 139 U.S.App.D.C. 60, 429 F.2d 739 (1970); Cf. United States v. Thompson, 483 F.2d 527 (3d Cir. 1973) cert. denied 415 U.S. 911, 94 S.Ct. 1456, 39 L.Ed.2d 496 (1974) (disqualifying judge because of “fixed view” that draft offenders must serve 30 months); United States v. Townsend, 478 F.2d 1072 (3d Cir. 1972) (semble). While this rule has been attributed to “enlightened policy rather than constitutional imperative”, Woodson, 428 U.S. 280, 303-305, 96 S.Ct. 2978, 2991, 49 L.Ed.2d 944 (1976), it has been argued that in situations where grave constitutional interests are at stake, “structural due process” requires individualized decision making. See Tribe, Structural Due Process, 10 Harv.Civ. Rts.-Civ.Lib.L.Rev. 269 (1975); Crawford v. Cushman, 531 F.2d 1114, 1125 (2d Cir. 1976).
. See Marshall v. United States, 414 U.S. 417, 94 S.Ct. 700, 38 L.Ed.2d 618 (1974) (upholding statute which denies drug rehabilitation sentence to two-time felons).
. C.F.R. 41329-30 (Sept. 5, 1975).
. Morrisey v. Brewer, 408 U.S. 471, 477, 92 S.Ct. 2593, 2598, 33 L.Ed.2d 484 (1972). See Gagnon v. Scappareli, 411 U.S. 778, 785, 93 S.Ct. 1756, 1761, 36 L.Ed.2d 656 (1973) (“the rehabilitative rather than punitive focus of the parole/probation system”); Zerbst v. Kidwell, 304 U.S. 359, 363, 58 S.Ct. 872, 874, 82 L.Ed. 1399 (1938) (“a means of restoring offenders who are good social risks to society”).
. Cf. United States v. Murray, 275 U.S. 347, 48 S.Ct. 146, 149, 72 L.Ed. 309 (1927) (“The parole statute provides a board to be invested with full opportunity to watch the conduct of penitentiary convicts.”)
. See, Ex Parte United States, 242 U.S. 27, 41, 37 S.Ct. 72, 61 L.Ed. 129 (1916) (“indisputably, under our constitutional system, the right to try offenses against the criminal laws and upon conviction to impose the punishment provided by law is judicial”); Yale Note supra note 2, at 892-893; cf. Wong Wing v. United States, 163 U.S. 228, 16 S.Ct. 977, 41 L.Ed. 140 (1896) (penalty of “hard labor” is an infamous punishment and therefore cannot be administratively imposed). North v. Russell, 427 U.S. 328, 96 S.Ct. 2709, 49 L.Ed.2d 534 (1976) (criminal trial before non-law trained judge is constitutional if trial de novo is available).
In view of this alteration in functions, prior cases upholding the federal parole statute against challenges based on alleged usurpation of judicial functions would not seem disposi-tive. E. g., Sims v. Rives, 66 U.S.App.D.C. 24, 84 F.2d 871, 879 (1936); see Commonwealth ex rel. Banks v. Cain, 345 Pa. 581, 28 A.2d 897, 143 A.L.R. 1473 (1942); Annotation, 143 A.L.R. 1726 (1942). Cf. Dreyer v. Illinois, 187 U.S. 71, 78-84, 23 S.Ct. 28, 47 L.Ed. 79 (1902) (Illinois indeterminate sentence act did not violate due process by vesting parole power in executive).
. See United States v. Brown, 381 U.S. 437, 441-447, 85 S.Ct. 1707, 1711, 14 L.Ed.2d 484 (1965) (separation of powers is at the root of Bill of Attainder clause, and “the legislative branch is not so well suited as politically independent judges and juries to the task of ruling upon the blameworthiness, of, and levying appropriate punishment upon specific persons”).
. Cf. Duncan v. Kahanamoku, 327 U.S. 304, 66 S.Ct. 606, 90 L.Ed. 688 (1946) (since Congress “did not explicitly declare” civil courts supplanted under martial law statute, Court rejected such authority in light of “principles and practices developed during the birth and growth of our political institutions.”); Soloway v. Weger, 389 F.Supp. 409, 411 (W.D.Pa.1974) (“blind adherence” to guidelines “may well be” intrusion into court’s sentencing role in excess of statutorily conferred powers of Parole Board); Kastenmeir & Eglit, Parole Release Decision-Making; Rehabilitation, Expertise and the Demise of Mythology, 22 Am.U.L.Rev. 477, 494, 495, 508, 522 (1973) (questioning assumption by Parole Board of judicial role of evaluating severity of offense in absence of legislative authorization). See generally Gewirtz, The Courts, Congress, and Executive Policy Making; Notes on Three Doctrines, 1976, Law and Cont. Prob. 46. But cf. Wiley v. United States Bd. of Parole, 380 F.Supp. 1194, 1196-97 (M.D. Pa. 1974) (upholding guidelines against challenge that they amounted to “resentences”); Manos v. United States Board of Parole, 399 F.Supp. 1103, 1105 (M.D.Pa.1975) (guidelines are “constitutional”). Garcia v. United States Bd. of Parole, 409 F.Supp. 1230, 1238-39 (N.D. filed 1976) revd on other grounds, 557 F.2d 100 (1972) (holding guidelines are not “usurpation” of sentencing function).
. See United States v. Bass, 404 U.S. 336, 348, 92 S.Ct. 515, 523, 30 L.Ed.2d 488 (1971) (“Because of the seriousness of penalties, and because criminal punishment usually represents the moral condemnation of the community, legislatures and not courts should define criminal activity”); Bell v. United States, 349 U.S. 81, 89, 75 S.Ct. 620, 622, 99 L.Ed. 905 (1955) (“The punishment appropriate for the diverse federal offenses is a matter for the discretion of Congress”); Ex Parte Grossman, 267 U.S. 87, 114, 45 S.Ct. 332, 335, 69 L.Ed. 527 (1925) (“The legislative authority of the Union must first make an act a crime, affix a punishment to it” before it may be punished); Ex Parte United States, 242 U.S. 27, 42, 37 S.Ct. 72, 74, 61 L.Ed. 129 (1916) (“indisputable also is that the authority to define and fix the punishment of a crime is legislative”); United States v. Wiitberger, 18 U.S. [5 Wheat.] 35, 43, 5 L.Ed. 37 (1820) (“the plain principle that the power of punishment is vested in the legislative not in the judicial department. It is the legislature not the court which is to define a crime and ordain its punishment”); Schick v. Reed, 419 U.S. 256, 274-76, 95 S.Ct. 379, 389, 42 L.Ed.2d 430 (1974) (“Prescribing punishment is a prerogative reserved for the lawmaking branch of government, the legislature”) (Marshall, J. dissenting).
. United States v. Bass, 404 U.S. 336, 348, 92 S.Ct. 515, 30 L.Ed.2d 488 (1971). Cf. Gut-knecht v. United States, 396 U.S. 295, 90 S.Ct. 506, 24 L.Ed.2d 532 (1970) (draft board is not authorized to punitively induct registrant who turned in registration card).
Unlike the case of Schick v. Reed, 419 U.S. 256, 95 S.Ct. 379, 42 L.Ed.2d 430 (1974), in which the Court upheld the imposition of conditions in connection with a presidential pardon, the parole board’s authority does not find its roots in the pardoning powers of Article II, but in specific legislative delegation. In 1927, seventeen years after the enactment of the federal parole statute, the Supreme Court undertook to survey the various forms of reduction of sentence available to prisoners. United States v. Murray, 275 U.S. 347, 48 S.Ct. 146, 72 L.Ed. 309 (1927) (construing probation act). It discerned “three different methods of mitigation:” the judicial probation authority granted by statute; “clemency by the President under the Constitution;” and “the power of parole by a Board of Parole abating judicial punishment.”
. See Yale Note supra note 2 at 888.
. In 41 Fed.Reg. 37316 (3 Sept. 1976), for example, the offense severity for immigration law violators was augmented as a result of the “Commission’s concern with the increasing numbers of immigration law violators in recent years.” Such a judgment appears to be one that is appropriate in the first instance for the Congress.
. It is clear that the “severity levels” in the parole guidelines do not correspond to the maximum punishments provided by “Congress. While uttering a counterfeit note carries a statutory maximum of 15 years (18 U.S.C. § 472 (1970)), if the currency passed does not exceed $1,000, the Commission classes the offense as “low moderate.” In contrast, mailing an anonymous threat to property, a violation whose statutory penalty is 2 years in jail (18 U.S.C. § 876 (1970)), is classified as “moderate” severity, requiring a longer sentence to be served before parole.
Illegally selling alcohol in the vicinity of Indian schools (18 U.S.C. § 1154 (1970)) and violation of the Mann Act (18 U.S.C. § 2421 et seq. (1970)) are classified as “low moderate” and “high” severity crimes respectively, despite the fact that they carry identical 5 year penalties. Misprision of a felony, (18 U.S.C. § 4 (1970)) and extortion by a United States employee (18 U.S.C. § 872 (1970)) are each punishable by a maximum of 3 years imprisonment, yet the former is rated as a “moderately” severe crime, and the latter is determined to be of “very high” severity.. Both carry lighter sentences than the Selective Service violations (50 App. U.S.C. § 462 (repealed) (5 year penalty)) which the Commission treats as “low-moderate” severity.
. Cf. Senate Bill 1437, 95th Cong. 1st Sess. (1977), Reported S. Rept. 95-605, 123 Cong. Rec.S 19037 (daily ed. Nov. 15, 1977) (providing for comprehensive code of criminal penalties and sentence review). Since S. 1437 would also largely eliminate parole, many of the difficulties raised by the Parole Commission’s modus operandi would, if S. 1437 is enacted, become truly moot.
. See National Cable Television Assn. v. United States, 415 U.S. 336, 94 S.Ct. 1146, 39 L.Ed.2d 370 (1974) (statute should not be construed to delegate taxing power without appropriate standards); Arizona v. California, 373 U.S. 546, 626, 83 S.Ct. 1468, 10 L.Ed.2d 542 (1963) (Harlan, J., dissenting) (statute should not be read to delegate power to allocate water without appropriate standards); L. Tribe, American Constitutional Law §§ 15-17, 284-291 (1978); Gewirtz, supra note 110, Wright, Book Review, Beyond Discretionary Justice, 81 Yale L.J. 575, 580, 582-87 (1972) (Wright, J.); Freedman, Review, Delegation of Power and Institutional Competence, 43 U.Chi.L.Rev. 307 (1976).
. Since Geraghty was sentenced under 18 U.S.C. § 4208(a)(2) (1970), he became eligible for parole immediately upon incarceration.
. 18 U.S.C. § 4203 (1970).
. Defendant’s Answer to Complaint ¶ 48. A39.
. 1 W. Blackstone, Commentaries *46.
. See Crosskey, The True Meaning of the Constitutional Prohibition of Ex Post Facto Laws, 14 U.Chi.L.Rev. 539 (1947); Note, Ex *264Post Facto Limitations on Legislative Power, 73 Mich. 1491, 1500-01 (1975).
. Art. I § 9 and § 10. For an excellent recent analysis of the ex post facto clause, see Tribe, supra note 117, 474-484 §§ 10-1-10-3.
. Calder v. Bull, 3 U.S. (3 Dallas) 385, 390, 1 L.Ed. 648 (1798) (opinion of Chase, J.). See United States Trust v. New Jersey, 431 U.S. 1, 10-11 n. 13, 97 S.Ct. 1505, 52 L.Ed.2d 92 (1977).
. Calder v. Bull, 3 U.S. [3 Dallas] at 390.
. Kring v. Missouri, 107 U.S. 221, 235, 2 S.Ct. 443, 455, 27 L.Ed. 506 (1882) quoting United States v. Hall, 26 Fed.Cas.No.15,285, p. 84, 2 Wash.C.C. 366, aff’d 10 U.S. [6 Cranch] 171, 3 L.Ed. 189 (1810) (repeal of provision that conviction of second degree murder operated as acquittal of first degree murder). The court in Kring went on to say: “No one can be criminally punished in this country, except according to a law prescribed for his government by the sovereign authorities before the imputed offense was committed, and which existed as a law at that time.”
. In re Medley, 134 U.S. 160, 10 S.Ct. 384, 33 L.Ed. 835 (1890) (new law mandating convicted murderer be kept in solitary confinement until execution and that warden should set date of execution without informing prisoner violated ex post facto clause).
. Lindsey v. Washington, 301 U.S. 397, 400, 57 S.Ct. 797, 798, 81 L.Ed. 1182 (1937).
. Lindsey was recently interpreted in Dob-bert v. Florida, 432 U.S. 282, 300, 97 S.Ct. 2290, 2301, 53 L.Ed.2d 344 (1977) as holding that *265“one is not barred from challenging a change in the penal code on ex post facto grounds simply because the sentence he received under the new law was not more onerous than that which he might have received under the old.”
. 417 U.S. 653, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974).
. 417 U.S. at 658-661, 94 S.Ct. 2532.
. 417 U.S. at 662-63, 94 S.Ct. at 2538.
. 277 F.Supp. 644 (D.Mass.1967) aff’d 390 U.S. 713, 88 S.Ct. 1409, 20 L.Ed.2d 150 (1968).
. 277 F.Supp. at 646.
. 390 U.S. 713, 88 S.Ct. 1409, 20 L.Ed.2d 150 (1968).
. This Court held in U. S. ex rel. Forino v. Garfinkel, 166 F.2d 887 (1948) that repeal of a provision removing civil disabilities upon completion of a sentence was not an ex post facto law. This holding has been substantially un*266dercut by the Supreme Court’s affirmance of Scafati. It has also been rendered inapplicable to this case by the recognition of parole and good time as protected liberty interests. Mor-rissey v. Brewer, 408 U.S. 471, 92 S.Ct. 2593, 33 L.Ed.2d 484 (1972) and Wolff v. McDonnell, 418 U.S. 539, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974).
We likewise decline to'follow, and question the continued viability of Singleton v. Shafer, 313 F.Supp. 1094 (E.D.Pa.1970) (repeal of good time statute not ex post facto) and Graham v. Thompson, 246 F.2d 805 (10th Cir. 1957) (repeal of good time statute not ex post facto, resting on interpretation of Utah law).
. 556 F.2d 648 (2d Cir. 1977).
. Id. at 654.
. 555 F.2d 1331 (6th Cir. 1977).
. Id. at 1335.
. Shepard v. Taylor, 556 F.2d at 654 (dictum).
. Ruip v. United States, 555 F.2d at 1335. Cf. Milhouse v. Levi, 179 U.S.App.D.C. 1, 548 F.2d 357 (1976) (alteration of eligibility regulations for furlough program not ex post facto law since furlough, unlike parole, was not part of the punishment).
. Love v. Fitzharris, 460 F.2d 382 (9th Cir. 1972) vacated as moot 409 U.S. 1100, 93 S.Ct. 896, 34 L.Ed.2d 682 (1973) (change in administrative interpretation of parole statute). See Marks v. United States, 430 U.S. 188, 191, 97 S.Ct. 990, 993, 51 L.Ed.2d 260 (1977) quoting Bouie v. Columbia, 378 U.S. 347, 353, 84 S.Ct. 1697, 1702, 12 L.Ed.2d 894 (1964) (“If a stats legislature is barred by the Ex Post Facto Clause from passing such a law, it must follow that the State Supreme Court is barred by Due Process from achieving the same result by judicial construction.”).
. 165 U.S.App.D.C. 284, 507 F.2d 1107 (1974).
. 165 U.S.App.D.C. 284 at 290, 507 F.2d 1107 at 1113.
. Conference Report at 36, 1976 U.S.Code Cong. & Admin.News, p. 368.
. 18 U.S.C. 4206(c) (1976).
. Conference Report at 27, 1976 U.S.Code Cong. & Admin.News p. 359.
. See Warden v. Marrero, 417 U.S. 653, 663, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974) (examining “practical effect” of parole limitations); Dob-bert v. Florida, 432 U.S. 282, 294, 97 S.Ct. 2290, 2298, 53 L.Ed.2d 344 (1977) (“We must compare the two statutory procedures in toto to determine if the new may be fairly characterized as more onerous”); Beazell v. Ohio, 269 U.S. 167, 170, 46 S.Ct. 68, 69, 70 L.Ed. 216 (1925) (“Just what alterations of procedure will be held to be of sufficient moment to transgress the constitutional prohibition cannot be embraced within a formula or stated in a general proposition. The distinction is one of degree. But the constitutional provision was intended to secure substantial personal rights against arbitrary and oppressive legislation.”); In Re Medley, 134 U.S. 160, 167-73, 10 S.Ct. 384, 33 L.Ed.2d 835 (1890) (analyzing practical effect of solitary confinement as increasing severity of punishment).
. Defendant’s answer to ¶ 48. A39.
. United States v. Salerno, 538 F.2d 1005, 1008 (1970). See Yale Note supra note 2, at 825 n. 75, 869 n. 293; United States v. Solly, 559 F.2d 230, 233 (3d Cir. 1977).
. The trial court should also examine the continued ability of Geraghty's counsel to represent such sub-classes adequately.