dissenting:
Ten years ago the Pennsylvania legislature enacted legislation1 which authorized county school boards to prepare, and the State Board of Education to approve, the reorganization and consolidation of school districts. Acting pursuant to that authority the Allegheny County school board prepar*1352ed, and the Pennsylvania State Board of Education approved, the reorganization which brought the General Braddock School District into being on July 1, 1971. On June 9, 1971, before the General Braddock School District ever came into existence, several parents brought an action under 42 U.S.C. § 1983 charging that statutory reorganization had created a segregated school system in which General Braddock School District was identifiably black and the surrounding districts were identifiably white. The complaint sought relief on behalf of named children of the plaintiff parents. It also sought class relief. On March 21,1972, the district court permitted the lawsuit to proceed as a class action. On May 15,1973, after an extensive trial, the district judge determined that the charge of intentional discrimination in the creation of the General Braddock School District had been proved, and ordered the prompt submission of proposals for interdistrict relief.
Today, seven years after suit was brought, and five years after the finding of de jure segregation, no remedy has been adopted. The majority opinion deals adequately with the doleful procedural posture of the case. Thus there is no need further to rehearse its history. The order appealed from, which effectively made an about face on the approach to a remedy for de jure segregation which the court had earlier proposed, was justified on five grounds: (i) a lack of evidence of broad based support for a plan of desegregation in the General Braddock community; (ii) failure of that community to propose any plan; (iii) the effect on non-white children of substantial busing to communities in which they would be identified as outsiders and would be unwelcome and a financial burden; (iv) the traffic problems involved in busing; and (v) loss of community control by the General Braddock community. In addition, the lower court proposed, under the imagined compulsion of Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974), to reopen the issue of whether school districts not created by Act 150 were active participants in the acts of deliberate discrimination charged in the complaint.
The majority, while it recites the procedural odyssey which produced the present state of inactivity, addresses none of these justifications for the lower court’s order because it holds that we lack appellate jurisdiction. I conclude that we have appellate jurisdiction. Because none of the reasons relied upon by the district court justify its extraordinary change of position, I would reverse the judgment appealed from.
An order which has the effect of denying all relief for a violation of a constitutional right is a final decision within the meaning of 28 U.S.C. § 1291. It is critical, therefore, to examine not only the verbal formula in the order appealed from, but the necessary effect of that order. Thus to some extent the finality and the propriety of the order are interrelated.
The first reason advanced by the district court for its about face on an interdistrict remedy is the absence of “broad based community support for such a remedy.” I would have thought that reliance upon community resistance to desegregation as a valid reason for denial of equitable relief was foreclosed as early as Cooper v. Aaron, 358 U.S. 1, 78 S.Ct. 1401, 3 L.Ed.2d 5 (1958). It is the district court’s obligation to order an effective solution for de jure segregation. Green v. County School Board, 391 U.S. 430, 439, 88 S.Ct. 1689, 20 L.Ed.2d 716 (1968). Community resistance, even when manifested by white flight from a school system, cannot justify a less than effective solution. Monroe v. Board of Commissioners, 391 U.S. 450, 459, 88 S.Ct. 1700, 20 L.Ed.2d 733 (1968). The district court found community resistance to an effective remedy six years after the suit commenced. That resistance is unlikely to diminish six or sixty years hence. If community resistance is recognized as a valid reason for denying interdis-trict relief, it can serve as a reason for denying such relief in perpetuity.
The district court also relied upon the failure of the affected community to propose a plan. The effect of this ruling, when combined with the court’s failure to order the state authorities to submit a new plan, *1353was to shift the burden of preparing a remedy for de jure segregation from the defendants, who were responsible for creation of the segregated district, and who have the technical and financial resources to remedy the violation, to the victims, who lack those resources.2 It is settled law that when intentional segregation has been found the responsibility for preparation of a remedial plan rests upon the school authorities and the district court, not upon the victims. E. g., Evans v. Buchanan (II), 582 F.2d 750, (3d Cir. 1978). Given the difficulty of preparing a satisfactory plan without the resources or expertise required to do so, the district court’s contrary ruling creates a substantial risk that the lawsuit will be relegated to a permanent deep freeze.
The third consideration relied upon by the district court was the fact that non-white children would be bused to communities where they would be unwelcome and a financial burden. That reasoning is simply a refinement of the court’s erroneous assumption that resistance to desegregation is a legally valid reason for the perpetuation of de jure segregation. No doubt there is unfortunate sentiment in the affected communities against busing black children to identifiably white schools and white children to identifiably black schools. Provisions in a plan designed to protect against adverse consequences of such sentiment are obviously appropriate. But to rely upon those feelings as a reason for denying relief legitimates them by suggesting that they are a permissible justification for governmental (in this instance judicial) inaction. So legitimated community sentiment stands as a perpetual barrier to the implementation of an effective remedy for de jure segregation. And insofar as tacit judicial acceptance of community opposition reinforces that opposition the barrier of community sentiment may well be higher hereafter than when the court first denied relief.3
In addition, the district court counted against the plan the fact that for General Braddock District it would have resulted in diminished control over its own schools. But that control exists only because of the alleged violation of the Fourteenth Amendment. If the passage of time has given the community a vested interest in control of the illegal district, each year that passes obviously will strengthen that interest, raising ever higher the bar to an effective remedy. Surely the district court’s reliance upon this factor was impermissible.
Finally, the district court expressed its belief that Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974) (Milliken I) “raises doubts about the application of an interdistrict remedy to districts that have not been full participants in all proceedings.” The court concluded, further, that the same districts must, under Milliken, be excluded from the remedy unless there is evidence of a deliberate, purposeful, segregative intent in their creation. Plainly this was a misreading of Milliken I. The limitations upon the equitable remedial powers of the federal courts outlined in that case clearly permit decrees reaching both districts which have committed acts of de jure discrimination, and districts in which significant segregative effects have occurred as a result of discriminatory acts committed elsewhere. Milliken v. Bradley, 418 U.S. at 744-45, 94 S.Ct. at 3127; accord, Hills v. Gautreaux, 425 U.S. 284, 293-94, 96 S.Ct. 1538, 47 L.Ed.2d 792 (1976). Moreover, Milliken reserved the question whether districts affected by a remedy must par*1354ticipate in litigation over the violation. 418 U.S. at 752-53, 94 S.Ct. 3112. This court in our prior opinion rejected the claim of potentially affected districts to be granted general intervention. Hoots v. Commonwealth of Pennsylvania, 495 F.2d 1095 (3d Cir.), cert. denied, 419 U.S. 884, 95 S.Ct. 150, 42 L.Ed.2d 124 (1974). In Evans v. Buchanan, supra, this court reviewed an interdistrict remedy in a case in which the affected districts did not participate in litigation of the alleged violation, although they had properly been allowed to intervene to present evidence bearing on the extent to which they had been affected by the violation. See Evans v. Buchanan, 393 F.Supp. 428, 431 (D. Del.), aff’d, 423 U.S. 963, 96 S.Ct. 381, 46 L.Ed.2d 293 (1975). The district court’s patent misapplication of Milliken I has the effect of starting this seven year old case all over again at the stage of litigation over violation, prolonging, perhaps for years, the denial of an effective remedy.
For me, this review of the effect of the lower court’s order clearly indicates that it is appealable. The immediately relevant precedent is Kelley v. Metropolitan County Board of Education, 436 F.2d 856 (6th Cir. 1970), which held that a district court order entered in August, 1970, staying all proceedings on proposed desegregation remedies for the Nashville school system in order to await the Supreme Court’s decisions in the 1971 school desegregation cases,4 was final under 28 U.S.C. § 1291. The majority appears to agree that the result in Kelley was sound. However, it finds that Kelley is distinguishable on two grounds. First, it is claimed that the order appealed from, unlike the order in Kelley, “interposes no bar to further proposals for relief.” Second, the majority argues that the order, by its terms, does not require that there be any more delay than may be necessary for the parties to formulate a plan.5
Whether one looks to the facts of Kelley or to its underlying rationale, the majority’s reasoning inadequately confronts the force of that opinion. It is highly formalistic to argue that the order entered here poses “no bar” to attainment of ultimate relief in this case. The effect of the order, as outlined above, is to place substantial, probably insuperable, obstacles in the way of the formulation of an effective decree. First, by rejecting the State defendants’ Plan A and the other plans submitted by local districts the order eliminates from consideration most if not all alternatives which might have been resorted to in order to remedy the interdistrict violation which has been found. Second, the trial court has relieved the state and county defendants, who committed the violation, and who alone have the financial resources and technical skills to devise a remedy, of the burden of doing so. While this order does not forbid the defendants from submitting new plans, it has shifted the task of formulating a plan to the plaintiffs, who lack the resources to do so. Third, the order places upon the plaintiffs the responsibility for generating community support, presumably both in and outside of the General Braddock district, for any plan they propose — a burden that is plainly illegal and which, based on the record before us, has thus far been beyond the capacity of even a well financed state agency to meet. Moreover, as each day passes the district court’s demand for a consensus plan becomes increasingly difficult to satisfy, since so long as white flight from the General Braddock district6 (which has in all probability been accelerated by the original segregation decision) continues, the mutual suspicion and mistrust between General Braddock and surrounding white districts can only increase. In my view, the order, as a practical matter, interposes a substantial “bar to further proposals for relief.”
*1355Second, while the order does not “by its terms” require that there be any more delay than may be necessary for the parties to formulate a plan, the prior history of this litigation indicates that even if the plaintiffs were able quickly to prepare and present a new plan, they could not expect a decision on the merits in less than two years. In view of the severe requirements which the court’s order imposed upon the plaintiffs, and the need, under the district court’s mistaken view of the case, for a partial relitigation of the merits, it is not at all likely that any new plan of relief will be rapidly formulated. Thus, assuming optimistically that plaintiffs could indeed formulate any plan meeting the standards of the lower court, the likelihood that such a plan would be implemented before the tenth anniversary of this action’s commencement, is, absent timely appellate intervention, remote. This two to three year delay contrasts sharply with the nine or ten month stay contemplated by the district court in Kelley.7
On the facts, then, this appeal presents a stronger case for a finding of finality than did the appeal in Kelley. More important, the majority’s argument against appealability wholly ignores the core of the Kelley opinion. As Judge Edwards wrote:
It is clear to us that the rights of school children to schooling under nondiscriminatory and constitutional conditions cannot be recaptured for any school semester lived under discrimination [sic] practices. Nor can any court thereafter devise an effective remedial measure. Therefore, we have no doubt that the District Court order of August 25, 1970, staying pupil integration proceedings for an indefinite time was final and appealable under 28 U.S.C. § 1291.
436 F.2d at 862. Precisely the same type of continuing, irremediable injury that served to justify the finding of finality in Kelley is present in this case. With the passage of each year, another class graduates from a segregated school system and is forever deprived of effective relief. For example, of the sixteen named student plaintiffs in this case, nine have already graduated from a segregated school system. 8 Five others will graduate by 1981,9 the year which, in light of the prior history of this case, we may view as a likely date for any new desegregation proposal. Eric Smith and Byron Knight, who were respectively in first grade and kindergarten in 1971, may, if they are lucky, see the implementation of some decree before they graduate from high school. Without appellate relief, the children of the students who graduated with the General Braddock School District’s first class in 1972 are likely to be attending segregated kindergartens in that district before any steps are taken to remedy the violation of law to which their parents were subjected.
Obviously the time demands of complex litigation make some delay in the provision of federal desegregation remedies inevitable, and the non-reviewability of most inter*1356locutory orders must be accepted as a necessary concomitant of that delay.10 But the lesson of Kelley is that when unconscionable delay threatens to work upon an entire generation of students a wholesale deprivation of constitutional rights,11 further needless extensions of the proceedings need not be tolerated. Such a deprivation is threatened here. In this case, there is in addition, a substantial likelihood that the conditions imposed by the lower court’s order will render impossible the formulation of an effective decree in any event — a factor not present in Kelley which also conduces strongly to a finding of finality.12
Read in this fashion, Kelley is consistent with the decisions of this court. Although this circuit has traditionally adopted a restrictive approach to finality, it has nevertheless allowed review of otherwise interlocutory orders where postponing review would effectively prevent vindication of an interest protected by federal law. In United States v. DiSilvio, 520 F.2d 247 (3d Cir.), cert. denied, 423 U.S. 1015, 96 S.Ct. 447, 46 L.Ed.2d 386 (1975), we held that a defendant in a criminal case could appeal from an order denying a motion to dismiss an indictment on double jeopardy grounds, because postponing review until after trial had the effect of destroying at least part of the substance of the constitutional rights being asserted. That approach to finality was approved in Abney v. United States, 431 U.S. 651, 97 S.Ct. 2034, 52 L.Ed.2d 651 (1977). Accord, United States v. Venable, No. 78-1224 (3d Cir. Sept. 22, 1978); United States v. Inmon, 568 F.2d 326 (3d Cir. 1977). In this case, for many class members, further delay threatens to destroy the entire right being asserted. It is no answer that in the dim and distant future other class members, perhaps still unborn, will be able to bring before us similar claims. There is no way that those plaintiffs can vindicate in any practical way the interests of litigants adversely affected by the district court order.
Similarly, when the interests of the federal government as a prosecutor are affected by a stay we have not hesitated to take a flexible approach to finality. In In re Grand Jury Proceedings, 525 F.2d 151 (3d Cir. 1975), the district court had stayed a federal grand jury investigation of criminal violations of the Clean Air Act in the steel industry pending the resolution of a parallel state court contempt proceeding involving the same alleged violations. This court held that the stay, though interlocutory, could be reviewed under § 1291 because the term of the grand jury might expire before the state court proceedings concluded, the statute of limitations continued to run, and violations of federal environmental statutes continued daily. Here the effect of the order appealed from, like the order held final in In re Grand July Proceedings, is to hamstring the effective vindication of federal interests. Like the statutory violations which evaded prosecution because of the running of the statute of limitations, the constitutional violations in this case cannot be remedied by later review. It is therefore difficult for me to perceive any princi*1357pled distinction between the posture in which the federal government found itself in In re Grand Jury Proceedings, and the plight of numerous class members in the instant case.
Three reasons usually are advanced for our resistance to entertaining appeals before the end of the entire case. The first is that interlocutory appeals often bring the legal issues to us in a posture in which they cannot be dealt with intelligibly. The second is that allowing interlocutory appeals produces delay. The third is our preoccupation with appellate caseloads. These reasons do not justify the dismissal of the instant appeal. I have found no difficulty in discussing, dissecting, and rejecting every reason relied upon by the district court in support of the order appealed from. On the record before us, where the very object of the appeal is to cure unconscionable delay which is causing irreparable harm, the majority’s reference to “the goal of speedy justice” as a justification for dismissing it rings hollow. And as to caseload considerations, this is one of those cases in which our concern about the caseload must yield to our obligation to enforce the supremacy of federal law. There is no justification for dismissing this appeal.13
CONCLUSION
As early as 1848 the Supreme Court stated that the statutory requirement of finality, when applied to cases seeking equitable relief, should not be read “in [a] strict and technical sense, but [must be] given . . . a more liberal, and as we think, a more reasonable construction, and one more consonant to the intention of the legislature.” Forgay v. Conrad, 47 U.S. (6 How.) 201,208, 12 L.Ed. 404 (1848). Chief Justice Taney recognized that when an interlocutory decree required the delivery of land and slaves from one party to another, the party deprived of their use and services was subjected to irreparable injury.
For the lands and slaves which they claim will be taken out of their possession and sold, and the proceeds distributed among the creditors of the bankrupt, before they can have an opportunity of being heard in court in defence of their rights. We think, upon sound principles of construction, as well as upon the authority of the cases referred to, that such is not the meaning of the acts of Congress.
Id. at 204. It is a nice irony that the final judgment rule could be construed to accommodate the interest of a master in the continued possession of his slave, but cannot, according to the majority, be construed to prevent the permanent loss to the class members in this case of the right to an education in a nonsegregated school.
I would vacate the judgment of the district court and would remand to the district court with directions to proceed forthwith with the selection and implementation of an interdistrict remedy submitted by the defendants. Given the majority’s position in this appeal, I can only suggest that if plaintiffs’ future efforts to obtain from the district court an order meeting our rigid tests for appealability prove ineffective, they should consider resort to our mandamus jurisdiction.
. Pub.L. 299, No. 150, Pa.Stat.Ann. tit. 24, § 2400.1 et seq. (Purdon Supp.1978), (“Act 150”).
. On December 23, 1974 the plaintiffs moved for the appointment of a consultant to prepare a plan for the court. To date no action has been taken on that ruling.
. Studies of plan acceptability conducted in 1974 and submitted to the district court suggest that public opposition will be the inevitable concomitant of any effective desegregation plan. According to the study, every plan proposed (except the “free choice” tuition plan, which was unacceptable for other reasons) faced substantial public opposition either in the surrounding communities or in General Braddock District. (App. at 350-352a). It is unlikely that this situation has changed since that time. Hence, it could be argued that the district court’s consensus requirement amounted to a holding that no relief would be allowed at all.
. E. g., Swann v. Board of Education, 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971).
. The majority also implies that the fact that Kelley had been pending for 15 years distinguishes the two cases. Majority Opinion, supra, at 1347 n.34. For the reasons set out, infra, at n.7,1 find this argument wholly unpersuasive.
. See, e. g., Hoots v. Commonwealth of Pennsylvania, 359 F.Supp. 807, 814-15 (1973).
. The delay contemplated here is not more excusable than that in Kelley because this case has “only” been pending for seven years. Kelley involved the desegregation of a district with 95,000 students and a long history of racial segregation. 436 F.2d at 859 n.4. While precise comparisons are not possible, the case here involves a total student population much smaller than that involved in Kelley, and there is, of course, no history prior to 1971 of deliberate segregation. Furthermore, in Kelley, once the actual constitutional violation had been adjudicated, the remedy followed within two years. See Kelley v. Metropolitan County Bd. of Educ., 463 F.2d 732 (6th Cir.), cert. denied, 409 U.S. 1001, 93 S.Ct. 322, 34 L.Ed.2d 262 (1972). That delay is dwarfed by the probable seven or eight year delay of remedy that could result from the holding of non-appealability in this case.
. In 1971, Ronald Knight was an eleventh grader. Loretta Knight was in tenth grade. Terrence Knight, Janelle Hoots and Tawanda Smith were all in ninth grade. Pamela Knight, Tevela Smith, and Joseph Smith were in seventh grade, and Lucinda Smith was a sixth grader. (App. at 17a-18a). Assuming that they were promoted in due course, all of these students have now graduated from high school.
. Darryl Knight (fifth grade); Robert Smith (fourth grade); Jamie Hoots, Marc Knight, and Wesley Smith (second grade). (App. at 17a-18a).
. This, I take it, is the thrust of the majority’s assertion that “ ‘the mere prospect of delay’ may not create appellate jurisdiction.” Majority Opinion, supra, at 1347.
. Cf. Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975), where Justice White listed as one of four exceptions to the rule of strict finality
situations where the federal claim has been finally decided, with further proceedings on the merits ... to come, but in which later review of the federal issues cannot be had, whatever the ultimate outcome of the case.
420 U.S. at 481, 95 S.Ct. at 1039. For the students who have graduated from the General Braddock District, the federal claim that was first raised in 1971 can no longer be reviewed, regardless of the outcome of this case.
. The majority suggests that it might in some other more compelling desegregation case be prepared to hold that an otherwise interlocutory order was a final denial of relief. Majority Opinion, supra, at 1346. Cf. Hart v. Community School Bd., 497 F.2d 1027, 1031 (2d Cir. 1974) (stating that “there may be circumstances where failure to order a desegregation long overdue constitutes a denial of an injunction even though those formal words were not uttered”). They do not, however, satisfactorily explain why this appeal does not present the proper occasion for such a holding.
. Since the order appealed from is final, and thus appealable under § 1291, it is unnecessary to join issue with the majority over its rejection of appellant’s alternative contention that the effect of the order was to modify an injunction. See 28 U.S.C. § 1292(a)(1).