[46] BORK, Circuit Judge, concurring in part and dissenting in part:

The majority holds that we have jurisdiction to hear this appeal, that appellants' ex post facto clause challenge is meritless, and that the record as it now stands is inadequate to decide either the statutory or equal protection issues. Accordingly, the majority remands the case to the district court for additional evidence. I agree with the first two holdings. From the third holding, I must respectfully dissent. In my view, no further factual development of the record is needed to see that, with the single exception of the sex discrimination claim, appellants' statutory and equal protection claims are meritless. I would therefore remand solely on the sex discrimination claim.

THE STATUTORY CLAIM

Appellants claim that D.C. Code § 24-209 (1981) forbids the disparate treatment of D.C. offenders on the basis of their place of incarceration — D.C. prisons or federal prisons. The majority declines to answer the question, remanding instead for a determination whether there is in fact disparate treatment. In my view, a trial is wholly unnecessary since the various sources for determining legislative intent — the language of the statute, its legislative history, its historical background, and its construction by the agency entrusted with its enforcement, see Howe v. Smith, 452 U.S. 473, 480-87, 101 S.Ct. 2468, 2473-77, 69 L.Ed.2d 171 (1981) — make clear that section 24-209 was not intended to prohibit the challenged disparate treatment.

Of the relevant sources for determining legislative intent, none supports appellants. The language of the provision, as the majority observes, excludes neither appellants' nor appellees' interpretation. The United States Parole Commission and the Attorney General are charged with enforcing section 24-209 in cases such as this; appellants have shown nothing to suggest that they have not consistently read the provision to permit application of federal parole standards to District offenders in federal prisons. A review of the historical background and legislative history of the statute, which must be the primary sources of illumination in this case, demonstrates that appellees' construction is the only one that makes sense as a conclusion about legislative intent.

Until the very end of the nineteenth century, no federal prisons existed, and federal prisoners were housed in state facilities. See Howe v. Smith, 452 U.S. at 483, 101 S.Ct. at 2475. The Supreme Court endorsed this practice. Ex parte Karstendick, 93 U.S. 396, 400, 23 L.Ed. 889 (1876). Beginning in 1789, the federal government asked the states to confine federal prisoners in their own prisons at the federal government's expense. Act of Sept. 23, 1789, 1 Stat. 96. Some states complied with this request. See McNutt v. Bland, 43 U.S. 9, 16, 2 How. 9, 16, 11 L.Ed. 159, modified, 43 U.S. 28, 2 How. 28, (1844) (discussing Mississippi statute); Randolph v. Donaldson, 13 U.S. 76, 84, 9 Cranch 76, 84, 3 L.Ed. 662 (1815) (discussing Virginia statute). In those states unwilling to accept federal prisoners, the First Congress and later Congresses permitted the federal marshal, under the direction of the federal district judge, to "hire a convenient place to serve as a temporary jail" until permanent arrangements could be made. Act of Mar. 3, 1791, 1 Stat. 225; Act of Mar. 3, 1821, 3 Stat. 646; Act of Mar. 2, 1833, ch. 57, § 6, 4 Stat. 632, 634. See Randolph v. Donaldson, 13 U.S. at 49, 9 Cranch at 84 (discussing 1791 Act).

Initially, federal courts were empowered by Congress to assign federal prisoners to individual prisons. Federal courts could sentence federal prisoners to any willing state prison in the state of conviction, inside or outside the district in which the court sat. Act of Mar. 3, 1825, ch. 65, § 15, 4 Stat. 115, 118; Act of Mar. 3, 1835, ch. 40, § 5, 4 Stat. 775, 777; Act of Mar. 28, 1856, ch. 9, 11 Stat. 2; Act of Mar. 3, 1865, ch. 86, 13 Stat. 500. During the Civil War, Congress authorized the Secretary of the Interior, in addition to federal courts, to assign federal prisoners to state prisons. Act of May 12, 1864, ch. 85, §§ 1 2, 13 Stat. 74, 74-75 (adult prisoners); Act of Mar. 3, 1865, ch. 121, §§ 1 2, 13 Stat. 538, 538 (juvenile offenders). A few years later, Congress gave the Attorney General broad authority to assign federal prisoners to any suitable jail or prison. Act of Mar. 5, 1872, ch. 30, § 1, 17 Stat. 35, 35; Act of July 12, 1876, ch. 183, 19 Stat. 88. The Supreme Court took note of the Attorney General's broad authority and held that a federal prisoner cannot object to the Attorney General's assignment. Ex parte Karstendick, 93 U.S. at 400-01; see United States v. Marshall, 17 D.C. (6 Mackey) 34, 37 (1887). See also Ex parte Waterman, 33 F. 29, 29-31 (N.D.N.Y. 1887) (transfer of prisoner by Attorney General can be done ex parte).

States that received federal prisoners were authorized by Congress to confine them for exactly the length of their sentence. In particular, state prison officials had no power, absent specific Congressional authorization, to release federal prisoners before the expiration of their term. See McNutt v. Bland, 43 U.S. at 8-10, 2 How. at 16-18. Otherwise, state officials had broad authority over federal prisoners. Indeed, federal law provided that

whenever any criminal convicted of any offence against the United States, shall be imprisoned, in pursuance of such conviction, and of the sentence thereupon, in the prison or penitentiary of any state or territory, such criminal shall in all respects be subject to the same discipline and treatment, as convicts sentenced by the courts of the state or territory, in which such prison or penitentiary is situated; and while so confined therein, shall also be exclusively under the control of the officers having charge of the same, under the laws of the said state or territory.

Act of June 30, 1834, ch. 163, 4 Stat. 739. See Mackin v. United States, 117 U.S. 348, 352, 6 S.Ct. 777, 779, 29 L.Ed. 909 (1886) (discussing 1834 Act); Ex parte Karstendick, 93 U.S. at 398, 399 (same).

By the end of the nineteenth century, most federal prisoners were concentrated in a few state prisons. Many states had found their prisons overcrowded with federal prisoners; some states had stopped accepting federal prisoners not convicted in their own state; and one state had refused to accept any federal prisoners at all. H.R. Rep. No. 1341, 61st Cong., 2d Sess. 7 (1910). In 1891, Congress authorized the construction of the first federal penitentiaries, Act of Mar. 3, 1891, ch. 529, 26 Stat. 839, and construction of Leavenworth prison began in 1896. H.R. Rep. No. 1341, supra, at 5, 6, 7.

Penal reform in the late nineteenth century encompassed the birth of parole, "a means of restoring offenders who are good social risks to society . . . under guidance and control" of a parole board. Zerbst v. Kidwell, 304 U.S. 359, 363, 58 S.Ct. 872, 874, 82 L.Ed. 1399 (1938). See 4 Attorney General's Survey of Release Procedures 4 (1939) (" Attorney General's Survey") (defining parole as "the release of an offender from a penal or correctional institution, after he has served a portion of his sentence, under the continued custody of the State and under conditions that permit his reincarceration in the event of misbehavior"). Like the closely allied but later developing practice of indeterminate sentencing, parole reflected the contemporary penological commitment to individualized treatment and was based at least as much on the idea of protecting society from newly released but still dangerous criminals as on the idea of offering rehabilitated offenders an early release. See National Commission on Law Observance and Enforcement, Report on Penal Institutions, Probation, and Parole 129-30 (1931) (Wickersham Report).

Informal methods of parole had existed since the New York House of Refuge had adopted an indenture system for juvenile delinquents in 1825. 4 Attorney General's Survey at 7. For adult offenders, the first modern attempt to institute a large-scale parole system began in 1876 at New York's Elmira Reformatory. Id. at 19. See The President's Commission on Law Enforcement and the Administration of Justice, Task Force Report: Corrections 60, 185 (1967). Only one year later, the New York legislature adopted the first American parole statute. 1877 N.Y. Laws, ch. 173, § 5. By 1910, thirty-one other states had also adopted parole laws, 4 Attorney General's Survey at 20, and Congress had adopted the first federal parole law, Act of June 25, 1910, ch. 387, 36 Stat. 819. 1 Attorney General's Survey at 15-16, 786.

Congress enacted the 1910 Federal Parole Act with two goals in mind, one penological, one fiscal. Believing that parole had "come to be recognized as humane, in the interest of sound policy and highly beneficial to the welfare of society," Congress intended federal prison officials to be able to reward rehabilitated prisoners as well as to maintain prison discipline through the incentive of an early release. H.R. Rep. No. 1341, supra, at 4. See id. at 4-5 (reprinting portions of 1898 American Bar Association report on parole). This policy was considered particularly appropriate for federal prisons because of the smaller number of prisoners who had committed serious offenses. Id. at 5 ("About 33 per cent of the prisoners in state penitentiaries are there for the commission of the smaller offenses, while about 57 percent of the prisoners in federal penitentiaries are there for the smaller offenses."). The second, fiscal purpose was to ease overcrowding in the three federal prisons, which were still only partially completed but which would remain overcrowded on completion as they received federal prisoners currently confined in state institutions.

To accomplish its goals in the existing system, the 1910 Act created separate parole systems for federal prisoners confined in federal and state institutions. Section 1 made every federal prisoner confined in a federal prison under a determinate sentence eligible for parole after having served one-third of his total term of incarceration. 1 Attorney General's Survey at 16. Section 2 created a separate parole board for each of the three federal prisons, and section 3 granted the boards the power to grant or revoke parole, subject to the approval of the Attorney General. 1 Attorney General's Survey at 16. Section 3 required the parole boards, in making parole decisions, to consider (1) whether "there is a reasonable probability" that the prisoner "will live and remain at liberty without violating the laws" and (2) whether "release is not incompatible with the welfare of society." 36 Stat. 819. Application of these nebulous standards by three separate federal parole boards undoubtedly led to disparate courses of decision in the three federal prisons. The Act left each parole board free to decide whether to release federal prisoners and what conditions of parole to prescribe. "The Board, under the law, establishes its own rules and regulations governing procedure, subject, of course, to the approval of the Attorney General," and aside from the two factors the Act required these boards to consider, "[a]ll other rules governing the condition of parole are left to the discretion of the Board and are promulgated in the rules laid down by them concerning their action and the conditions under which prisoners are released." White, The Federal Parole Law, 12 A.B.A.J. 51, 51 (1926) (L.C. White, Superintendent of Federal Prisons).

By contrast, the Act left federal prisoners confined in state prisons subject to state parole laws. Section 9 made such federal prisoners "eligible to parole on the same terms and conditions and by the same authority and subject to recommittal for violation of such parole in the same manner" as state prisoners. 36 Stat. 821; see H.R. Rep. No. 1341, supra, at 3; White, supra, 12 A.B.A.J. at 52. Section 9 explicitly adopted the state law for each federal prisoner confined in a state prison or reformatory. The only qualification was that all paroles for such prisoners were subject to the approval of the Attorney General. H.R. Rep. No. 1341, supra, at 3; White, supra, 12 A.B.A.J. at 52.

It is against this background of federal prison practices that District of Columbia law and practices must be understood. Until 1932, District of Columbia prisoners were subject to the jurisdiction of the federal parole system, and "the history of parole in the District of Columbia prior to 1932 is substantially that of the federal parole system." 1 Attorney General's Survey at 232 (footnote omitted). In particular, District prisoners confined in the District penal and reformatory institutions were considered for parole under the 1910 Federal Parole Act. Id. In 1932, Congress changed that situation by passing a statute entitled "An Act to establish a Board of Indeterminate Sentence and Parole for the District of Columbia and to determine its functions, and for other purposes." Act of July 15, 1932, ch. 492, 47 Stat. 696 (currently codified at D.C. Code § 24-201a et seq. (1981)) (the "District Parole Act"). The 1932 District Parole Act created a separate, local Board of Indeterminate Sentence and Parole to alleviate the burden placed on the Federal Parole Board by the need to make parole decisions for District as well as federal prisoners. H.R. Rep. No. 881, 72 Cong., 1st Sess. 2 (1932); S.Rep. No. 450, 72d Cong., 1st Sess. 2-3 (1932); see J. Nolan, One Hundred Prisoners: A Study of the Operation of Parole in the District of Columbia xxxiii (1936). See also H.R. Rep. No. 1341, supra, at 6 ("About 20 per cent of the prisoners at Leavenworth and Atlanta [as of 1910] are sent from the District of Columbia for violation of the laws of the District.").

The local Board of Indeterminate Sentence and Parole was authorized to parole prisoners held in District custody after making the same two inquiries as the 1910 Federal Parole Act required federal parole authorities to make. The discretion exercised by the District Board was also similar to that of the Federal Board. The members of the District Board, noted one contemporary observer, had "the license to make their own rules and regulations, and to determine the policies the board will follow." J. Nolan, supra, at 67. Compare White, supra, 12 A.B.A.J. at 51, quoted at pp. 1137-1138 supra, In practice, the Board considered a variety of factors in deciding whether to release a prisoner — among them, the prisoner's physical condition, his opportunity for employment, his institutional behavior, his likely response to parole supervision, his likely behavior after release, his prior criminal record, public opinion, and public safety. See J. Nolan, supra, at 53-77.

There were three important differences between the federal and District parole systems in 1932. The first and most obvious one was geographic. The Federal Parole Board had nationwide authority over all prisoners confined in federal prisons. The District Board had authority only over prisoners in District custody. Second, federal parole authorities had jurisdiction only over federal prisoners serving definite terms of imprisonment or life terms. See White, supra, 12 A.B.A.J. at 51. As indeterminate sentencing did not exist at that time for federal offenders, all federal prison terms fit into one of these two categories. See 4 Attorney General's Survey at 20 n. 41. The District Parole Board, unlike its federal counterpart, had jurisdiction over prisoners serving indeterminate terms, since as a result of the 1932 District Parole Act, all District offenders received indeterminate sentences — minimum and maximum terms of confinement set by the sentencing judge, with the District parole authorities determining the actual period of incarceration. H.R. Rep. No. 881, supra, at 2-3; S.Rep. No. 450, supra, at 3. Third, prisoners confined in federal institutions became eligible for parole after serving one-third of their total term of incarceration. District prisoners confined in District custody became eligible for parole after serving only one-fifth of their total term. S.Rep. No. 450, supra, at 3; 75 Cong.Rec. 15,035-37 (1932) (remarks of Rep. Black). The 1932 Act thus created three important differences in the treatment of prisoners in federal custody and those in District custody. Nothing in the Act, however, modified the different parole treatment federal prisoners received in state custody.

The statute at issue in this litigation, section 24-209 of the current District of Columbia Code, was not part of the original District Parole Act. It was added as an amendment in 1934. The amendment granted federal authorities "the same power and authority" over District prisoners in their custody as the District Board would have if these prisoners were in District custody. The legislative history of this amendment is sparse: it consists primarily of a letter from M.C. Hazen to the Senate and House Committees on the District of Columbia, a letter that, with a few introductory remarks, makes up the entire Senate and House Reports. See S.Rep. No. 674, 73d Cong., 2d Sess. 1-2 (1934) (reprinting letter); H.R. Rep. No. 1446, 73d Cong., 2d Sess. 1-2 (1934) (same). Hazen wrote that the Commissioners of the District had submitted the proposed amendment in order to rectify a problem existing at Lorton, the District reformatory for juvenile offenders. When Congress enacted the District Parole Act in 1932, according to Hazen, "it was the intention to erect a penitentiary upon a portion of the site occupied by the reformatory at Lorton, Va., in which to confine hardened criminals for whom the conditions at the reformatory were unsuitable." H.R. Rep. No. 1446, supra, at 1-2 (Hazen). See J. Nolan, supra, at xxviii. When that project was later abandoned for lack of money, it became necessary to transfer District prisoners to federal prisons. This transfer created a problem: the Federal Parole Board had jurisdiction to parole inmates only if they were serving definite terms of incarceration or life sentences, but all District offenders were given indeterminate sentences. H.R. Rep. No. 1446, supra, at 2. The Federal Parole Board therefore had no power to release District prisoners even though Congress had contemplated, in passing the 1932 District Parole Act, that District offenders would be eligible for parole. The amendment Hazen proposed was necessary to ensure that the Federal Parole Board had authority to parole District offenders who were transferred to federal prisons. Id.

On April 25, the Senate passed the amendment without debate and sent it to the House. 78 Cong.Rec. 7276 (1934). The House approved its own version of the Senate bill without debate on May 28. 78 Cong.Rec. 9724 (1934). Representative Jenckes was the only person to comment upon the House bill. He offered these comments:

Briefly summed up this measure permits the authorities at Lorton to transfer hardened criminals or unruly convicts to other Federal prisons. In this manner the objective of Lorton can be maintained.

78 Cong.Rec. 9745 (1934). After praising the superintendent of prisons, Rep. Jenckes went on:

I believe that the objectives and ideals of the Lorton institution should have the full and sympathetic support of the Congress, and I hope that my colleagues will enact this measure without delay, as it will increase the efficiency of one of America's penal institutions.

Id. On May 31, the House unanimously voted to substitute the Senate version for its bill. 78 Cong.Rec. 10,137 (1934). The bill was then presented to President Roosevelt who signed it into law on June 5. 78 Cong.Rec. 10,769 (1934).

In light of the historical background set out above, the 1934 amendment to the District Parole Act must be read as having had a very limited purpose. Congress was fully aware that United States prisoners could be confined in institutions under three separate jurisdictions (federal, District, or state), that the Attorney General had the authority to transfer United States prisoners across jurisdictional lines, and that a prisoner's eligibility for parole was governed by the rules of the jurisdiction having custody over him. Under existing law, federal prisoners could be subject to any one of three different types of parole systems, and to any of a much larger number of particular systems: at least 32 states had adopted parole by this time. (In addition, prior to the consolidation of federal parole authority in 1930, federal prisoners in federal custody could be subject to any of the different parole release standards for the several federal prisons.) Congress assumed that transferring a prisoner from one jurisdiction to another subjected that prisoner to the standards and authority of the parole board of the receiving jurisdiction. If, as appellants argue, the purpose of the 1934 amendment was to make the Federal Parole Board an agent of the District Board, Congress would surely have noted that the problem of parole standards depending on location of confinement was not restricted to the class of District offenders in federal prisons. Congress might even have altered the Attorney General's plenary transferring authority. But Congress did no such thing.

Other than the 1932 Act, Congress modified none of the statutes that gave rise to different standards for release. Nothing in the language or legislative history of the 1934 amendment even indicates Congressional awareness of any fairness issue raised by the transfer problem. There is simply no basis for believing that in enacting the 1934 amendment Congress was making a radical departure from its long and consistent practice of permitting parole standards to vary from custodial jurisdiction to custodial jurisdiction without regard to the jurisdiction of conviction. Rather, the only plausible construction of the statute is that it was designed to return to the Federal Parole Board a part of its jurisdiction that had been eliminated by the indeterminate sentencing provisions of the 1932 Act. Thus, the 1934 amendment simply ensured that a particular change in custody did not result in the denial of any opportunity for parole altogether. Indeed, it did not even do quite that: by leaving intact the Attorney General's authority to transfer a District prisoner from a federal prison to a state institution — an authority ratified by Congress in 1930, Act of May 14, 1930, ch. 274, § 4, 46 Stat. 325, 326; see S.Rep. No. 533, 71st Cong., 2d Sess. 1, 3 (1930); H.R. Rep. No. 106, 71st Cong., 2d Sess. 3 (1930) — the 1934 amendment guaranteed only that a District prisoner would be eligible for release on parole if he remained in federal custody, for the state law applicable to prisoners in state custody might not provide for parole.

Appellants offer an alternative interpretation of section 24-209. They argue that Congress adopted the provision in order to reverse the decision of the Fifth Circuit in Lee v. Aderhold, 5 F.Supp. 950 (N.D.Ga. 1933), rev'd, 68 F.2d 824 (5th Cir.); cert. denied, 292 U.S. 633, 54 S.Ct. 718, 78 L.Ed. 1486 (1934). The Fifth Circuit held in that case that District prisoners confined in federal prisons could be considered for parole under federal, not District, guidelines, and it rejected a prisoner's argument that this disparity violated the equal protection clause. Aderhold v. Lee, 68 F.2d at 825. The majority suggests, but does not conclude, that this interpretation of section 24-209 is correct. Majority op. 1138. I believe that it is unsupportable.

The legislative history is totally silent on the existing disparities in the parole laws created by the 1910 Federal Parole Act and the 1932 District Parole Act. M.C. Hazen, the proponent of section 24-209, did not refer to these disparities or to Aderhold at all in his letter to the Senate and House Committees on the District of Columbia. S.Rep. No. 674, supra, at 1-2 (reprinting letter); H.R. Rep. No. 1446, supra, at 1-2 (reprinting letter). Neither the Senate Report nor the House Report nor Representative Jenckes — the only sources of legislative history for the provision — commented on these disparities or on the Aderhold decision. See pp. 1139-1140 supra. There is no basis whatsoever for the conclusion that Congress intended to overrule the decision in Aderhold: it is nowhere mentioned in the legislative history, and indeed there is no evidence that Congress was even aware of it.

On other occasions, Congress has clearly manifested its intention to overturn a decision, even in the field of parole law, with which it has disagreed. For example, when Congress enacted the Act of Oct. 26, 1974, Pub.L. No. 93-481, § 2, 88 Stat. 1455, overruling Warden v. Marrero, 417 U.S. 653, 94 S.Ct. 2532, 41 L.Ed.2d 383 (1974) (repeal of statute that denied parole to certain narcotics offenders did not make previously convicted drug offenders eligible for parole), the legislative history expressly indicated the intent to overrule the case. H.R. Rep. No. 1442, 93d Cong., 2d Sess. 6-7 (1974); H.R. Rep. No. 1248, 93d Cong., 2d Sess. 8 (1974), U.S. Code Cong. Admin. News 1974, p. 5971. It is impossible to believe that if so short a statute as section 24-209 had so clear a purpose as to overrule Aderhold the legislative history would nowhere have mentioned the case. Indeed, assuming that Congress was aware of Aderhold, its failure to mention the case, when enacting a statute that is at best ambiguous on the issue decided in the case, might more easily be read as an approval of the holding than as a disapproval of it. Cf. Merrill Lynch, Pierce, Fenner Smith, Inc. v. Curran, 456 U.S. 353, 380-89 nn. 65-66, 102 S.Ct. 1825, 1840-44 nn. 65-66, 72 L.Ed.2d 182 (1982) (Congress's recognition of judicial decisions implying a private right of action under the federal commodities trading laws in an amendment to those laws constitutes Congressional acceptance of those interpretations as correct).

If section 24-209 did not expressly repeal the Federal Parole Board's authority to apply its own standards, neither did it do so by implication. As a general rule, "repeals by implication are not favored," Posadas v. National City Bank, 296 U.S. 497, 503, 56 S.Ct. 349, 352, 80 L.Ed. 351 (1936), and "[i]n the absence of some affirmative showing of an intention to repeal, the only permissible justification for a repeal by implication is when the earlier and later statutes are irreconcilable." Morton v. Mancari, 417 U.S. 535, 550, 94 S.Ct. 2474, 2482, 41 L.Ed.2d 290 (1974). Section 24-209, as shown above, is readily reconcilable — indeed, is best read as being perfectly consistent with — the existing federal and District parole laws. The provision therefore cannot be read as repealing these laws by implication.

Appellants also seek support for their interpretation in Bracey v. Hill, 11 F.Supp. 148 (M.D.Pa.), aff'd, 77 F.2d 970 (3d Cir. 1935). In that case, a District prisoner transferred to a federal penitentiary in Georgia contended that he could not be transferred to a prison outside of the District of Columbia because he had been prosecuted, convicted, and sentenced under the District penal code. 11 F.Supp. at 149. The Court rejected the contention.

The only statement in Bracey that supports appellants is the statement that section 24-209 was enacted "[p]robably as a result of" Aderhold, a statement that cannot fairly be read as anything more than conjecture. 11 F.Supp. at 149. The Bracey court examined none of the background or legislative history of the section and cited no support for this guess at its origin. The remainder of Bracey, to the extent it is relevant, actually undermines appellants' argument because it recognizes the authority of the Federal Parole Board over District prisoners once they had been transferred from the District to a federal institution.

In any event, Bracey's comments on the 1934 statute were unnecessary to its decision. By confusing the Attorney General's authority to transfer custody of District prisoners with the Federal Parole Board's jurisdiction to release District prisoners, Bracey misinterpreted the 1934 amendment. That amendment was irrelevant to Bracey's claim, which challenged only the authority of the Attorney General to effect his transfer from the District of Columbia to Georgia. That claim had been meritless since Ex parte Karstendick, 93 U.S. at 404, and was not given life by either the 1932 District Parole Act or the 1934 amendment to that Act.

THE EQUAL PROTECTION CLAIMS

Appellants challenge the application of federal parole standards to them on two grounds under the due process clause of the Fifth Amendment. Their first claim — that equal protection is denied by treating District offenders held in District custody under allegedly easier parole standards than are applied to appellants as District offenders held in federal custody — is without merit as a matter of law. No point will be served, therefore, by the trial the majority requires on remand.

In Koyce v. United States Board of Parole, 113 U.S.App.D.C. 152, 306 F.2d 759 (1962), this court rejected an equal protection argument indistinguishable from that made by appellants. The discrimination alleged was that military prisoners released from civilian federal prisons were subjected to parole requirements not imposed upon military prisoners released from military prisons. Id. at 762. The court held that, "[i]n determining whether [a prisoner] is being denied equal protection of the laws the class to which he belongs consists of the persons confined as he was confined, subject to the same conditions to which he was subject." Id. The court summarily disposed of the equal protection claim, first, by recognizing that Congress had validly determined that it was "desirable and feasible" for all persons confined in the same federal institution to be subject to the same release conditions and, second, by expressly adopting the "reasoning and conclusion" of the Fifth Circuit Court of Appeals in Bates v. Wilkinson, 267 F.2d 779 (5th Cir. 1959). This court's express adoption of Bates's reasoning is significant for this case because Bates, in rejecting the same equal protection challenge as that made in Koyce, relied almost exclusively on its earlier decision in Aderhold v. Lee, 68 F.2d 824, and on Ex parte Karstendick, 93 U.S. 396, 23 L.Ed. 889. Aderhold had rejected the equal protection claim made here by appellants.

Appellants' first equal protection claim is thus foreclosed by an earlier decision of this court. Even if the issue were open, however, the claim should be dismissed. A discrimination based on place of custody must be upheld if it meets the rationality standard of equal protection review. That is made clear not only by Koyce but by the more recent Supreme Court case of McGinnis v. Royster, 410 U.S. 263, 93 S.Ct. 1055, 35 L.Ed.2d 282 (1973). In McGinnis, the Court rejected an equal protection challenge to a part of New York's parole scheme. "The determination of an optimal time for parole eligibility," said the Court, "elicited multiple legislative classifications and groupings, which the court below rightly concluded require only some rational basis to sustain them." The Court explained further: "We do not wish to inhibit state experimental classifications in a practical and troublesome area, but inquire only whether the challenged distinction rationally furthers some legitimate, articulated state purpose." Id. at 270, 93 S.Ct. at 1059. The Court found the New York scheme lawful because it rationally tied parole eligibility to the need and opportunities for rehabilitation of prisoners in different institutions.

The distinction appellants challenge here easily satisfies the rationality requirement. For most of our history, Congress has subjected prisoners to the parole standards of the jurisdiction operating the prison of confinement. That fact is itself legally significant. " 'If a thing has been practiced for two hundred years by common consent, it will need a strong case' to overturn it." Schick v. Reed, 419 U.S. 256, 266, 95 S.Ct. 379, 385, 42 L.Ed.2d 430 (1974) (quoting Jackman v. Rosenbaum Co., 260 U.S. 22, 31, 43 S.Ct. 9, 10, 67 L.Ed. 107 (1922) (Holmes, J.)).

The nature of a parole is significant as well. Parole, like confinement in any particular prison, is neither a constitutional or statutory right nor an intrinsic part of the sentence received on conviction. Rather, it is a matter committed by Congress to the discretion of the Executive Branch (as constrained by statute) to adopt measures for the effective administration of prisons, for the execution of just punishment, for the protection of the public, and for the rehabilitation of offenders. See Connecticut Board of Pardons v. Dumschat, 452 U.S. 458, 463-64, 101 S.Ct. 2460, 2463-64, 69 L.Ed.2d 158 (1981); Greenholtz v. Inmates of the Nebraska Penal and Correction Complex, 442 U.S. 1, 7, 99 S.Ct. 2100, 2103, 60 L.Ed.2d 668 (1979); Meachum v. Fano, 427 U.S. 215, 224, 96 S.Ct. 2532, 2538, 49 L.Ed.2d 451 (1976); Warren v. United States Parole Commission, 659 F.2d 183, 189-97 (D.C. Cir. 1981). As this court recognized in Koyce, Congress could rationally conclude that prisons should be managed largely on a prison-by-prison basis, allowing broad authority within each prison to set its own rules. Uniformity of parole standards within a prison might well be adopted as an important tool of prison management. Congress could rationally conclude that it would be deleterious to prisoner morale, discipline, and rehabilitation if persons confined within the same prison, perhaps within the same cell, were subject to different standards for release on parole.

In the area of parole practices, "few certainties exist." Greenholtz v. Inmates of the Nebraska Penal and Correctional Complex, 442 U.S. at 8, 99 S.Ct. at 2104. It is an area of experimentation, and actual practices reflect the accommodation of many policies. The rationality of the scheme challenged here thus would not be undermined even if appellants could show, as they have alleged, that appellees do not rigidly adhere to a policy of uniformity of parole standards within each prison. In particular, that the same policy may not be applied to women prisoners might well reflect a reasonable relaxation of the uniformity policy in order to accommodate another goal, that of according female District offenders parole treatment comparable to that received by most male District offenders. See pp. 1145-1146 infra. Moreover, even if state offenders in federal custody receive parole consideration by state authorities, as appellants allege, that practice would likewise not undermine the rationality of the scheme challenged here: the practice might reflect the special circumstances under which federal prisons take custody of state offenders or federal government respect of state sovereignty or other valid policies.

In sum, "there are plausible reasons for Congress' action" here, and it is " 'constitutionally irrelevant whether this reasoning in fact underlay the legislative decision.' " United States Railroad Retirement Board v. Fritz, 449 U.S. 166, 179, 101 S.Ct. 453, 461, 66 L.Ed.2d 368 (1980) (quoting Flemming v. Nestor, 363 U.S. 603, 612, 80 S.Ct. 1367, 1372, 4 L.Ed.2d 1435 (1960)). Because "the drawing of lines that create distinctions is peculiarly a legislative task and an unavoidable one[,] . . . [s]uch action by a legislature is presumed to be valid." Massachusetts Board of Retirement v. Murgia, 427 U.S. 307, 314, 96 S.Ct. 2562, 2567, 49 L.Ed.2d 520 (1976) (footnote omitted). Appellants have shown nothing that casts doubt on the rationality of Congress's tying parole standards for District offenders to place of custody, and hence that practice is valid.

United States v. Thompson, 147 U.S.App.D.C. 1, 452 F.2d 1333 (1971), does not suggest a contrary conclusion or a higher standard of review. The court there said that equal protection would be denied by application to persons charged with criminal violations of federal law in the District of Columbia of a higher bail standard than was applied to persons charged with the same crime in federal courts outside the District. That difference, according to the court, would carve out a special exception for the District from an otherwise uniform national law. The court found no rational basis for the exception. It stated that, in any event, more than mere rationality would be required to uphold the difference, both because a fundamental liberty interest was involved and because the scheme discriminated against the District of Columbia, which, having no representation in Congress, is deprived of the usual political check on discriminatory treatment.

Section 24-209 of the D.C. Code, construed to permit application of federal parole standards to persons convicted of local District crimes who are confined in federal prisons, has none of the characteristics that give rise to constitutional problems, and to special scrutiny, in Thompson. First, the statute does not make a special exception for the District from an otherwise uniform national law; rather, it is part of a uniform, and rational, national scheme that generally treats prisoners in federal custody according to federal parole rules. Second, the discrimination challenged here is not against District residents, not, that is, between persons outside the District and those within; rather, it is between two classes of persons convicted in the District, between local offenders in local custody and local offenders in federal custody. Section 24-209 thus is like any law governing the District that classifies and therefore does not call forth Thompson's suspicion of discrimination against a class unrepresented in the legislature. Finally, the fundamental liberty interest in Thompson is that of someone whose conviction may be reversed on appeal. Here, by contrast, the validity of the criminal conviction is not at issue, and whatever rights appellants may have regarding parole are fully respected by a rationality standard of review. McGinnis v. Royster, 410 U.S. 263, 270, 93 S.Ct. 1055, 1059, 35 L.Ed.2d 282 (1973).

Appellants' second equal protection claim is that application of federal parole standards to them, as male District offenders in federal custody, unconstitutionally discriminates against them because female District offenders in federal custody are considered for parole by the District parole board under District standards. I conclude that this claim, unlike appellants' other claims, warrants a remand because neither the legal nor the factual issues are sufficiently developed.

The alleged sex differential arises, somewhat ironically, from a consent decree in Garnes v. Taylor, Civ. No. 159-72 (D.D.C. Dec. 10, 1976), which sought to end an alleged discrimination in parole standards against female prisoners. The discrimination, if there was one, existed because there are no District prisons for long-term female offenders, all of whom are therefore confined in federal prisons and were subject to federal parole standards. Plaintiffs' theory was that this created gender discrimination. The Garnes consent decree provided for treating female District offenders (including U.S. Code violators) like the bulk of male District offenders, those in District custody, and unlike the small number of male District offenders in federal custody. This theory was not ruled upon, and the settlement appears to have been a governmental attempt to aid female prisoners previously disadvantaged because of their sex. It may be odd that the result should be that males in federal prisons, who previously had no valid equal protection argument because of parole standards more stringent than those applied to males in District custody, should acquire an equal protection claim because of an attempt to aid female prisoners. However that may be, the peculiar genesis and structure of the alleged present gender difference in parole treatment would seem to complicate the required constitutional analysis. Appellate decision would be greatly aided by a district court decision on the relevance of this history as well as the proper definition of the classes for equal protection analysis and whether any other justification for the alleged discrimination meets the applicable constitutional standard.

The sex discrimination issue is unclear on factual grounds as well. For appellants to make out their claim, they must, as a first step, prove differential treatment of men and women. The record does not currently contain that showing. Indeed, there is a strong suggestion in the record that, despite the Garnes decree, women District offenders are not consistently receiving parole consideration by the District board under District parole standards. See L. Steinitz, The Garnes Decree in Reality: Parole Eligibility and Determination for D.C. Women in Federal Correctional Institutions (1981), Appendix for Appellants at 124. In addition, it is not clear whether application of whatever standards are applied has resulted in treatment of men less favorable than that accorded women. At this time, in sum, it is impossible to say whether there is any significant difference in parole standards applied to males and females, let alone whether any difference there may be is unconstitutional. A remand on this issue is therefore necessary.