For the fourth time this court is called upon to review the protracted proceedings *245of this action brought by plaintiffs1 to obtain relief from alleged unconstitutional segregation of the Dayton public schools resulting from actions by defendants.2 Reference is made to the previous decisions of this court for a detailed recitation of facts and issues. See Brinkman v. Gilligan, 539 F.2d 1084 (6th Cir. 1976) (Brinkman III), vacated and remanded sub nom., Dayton Board of Education v. Brinkman, 433 U.S. 406, 97 S.Ct. 2766, 53 L.Ed.2d 851 (1977); Brinkman v. Gilligan, 518 F.2d 853 (6th Cir. 1975) (Brinkman II); Brinkman v. Gilligan, 503 F.2d 684 (6th Cir. 1974) (Brinkman I).
In its initial opinion filed February 7, 1973, the district court found that racially imbalanced schools, optional attendance zones, and the rescission by the Dayton Board of Education (hereinafter Board) of three resolutions calling for racial and economic balance in each public school were “cumulatively in violation of the Equal Protection Clause” of the Constitution. In Brinkman I, supra, 503 F.2d 684, this court affirmed the holding of the district court that the Dayton public schools were unlawfully segregated by race and also reviewed four school practices3 which allegedly maintained and expanded the segregated school system. This court determined that at that time it was unnecessary to consider whether these four practices should be included as part of the constitutional violation in view of the conclusion that the remedy ordered by the district court was inadequate “considering the scope of the cumulative violations.” Id. at 704.
Following remand, this court again rejected the desegregation plan adopted by the district court on the grounds that the plan failed to eliminate the “basic pattern of one-race schools” and the “continuing effects of past segregation” throughout the Dayton school system. Brinkman II, supra, 518 F.2d at 857. We again remanded the case to the district court with the following instructions:
On remand we direct that the court adopt a system-wide plan for the 1976-77 school year that will conform to the previous mandate of this court and to the decisions of .the Supreme Court in Keyes [Keyes v. School District No. 1, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548] and Swann [Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554]. We direct that this plan be adopted not later than December 31, 1975, so that it may be placed in effect at the beginning of the new school year in September 1976. Id. at 857.
After evidentiary hearings and the appointment of a master, the district court ordered the implementation of a systemwide desegregation plan for the 1976-77 school year subject to flexible guidelines.4
*246In Brinkman III, supra, 539 F.2d 1084, this court approved the systemwide plan which thus became operative for the 1976-77 school year. Subsequently, the Supreme Court vacated the judgment5 of this court and ordered that the case be remanded to the district court for further proceedings. Dayton Board of Education v. Brinkman, supra, 433 U.S. 406, 97 S.Ct. 2766, 53 L.Ed.2d 851 (1977). The Supreme Court directed that the district court:
first determine whether there was any action in the conduct of the business of the school board which was intended to, and did in fact, discriminate against minority pupils, teachers, or staff.
* * * * * *
If such violations are found, the District Court in the first instance, subject to review by the Court of Appeals, must determine how much incremental segre-gative effect these violations had on the racial distribution of the Dayton school population as presently constituted, when that distribution is compared to what it would have been in the absence of such constitutional violations. The remedy must be designed to redress that difference, and only if there has been a system-wide impact may there be a systemwide remedy, (citations omitted). 433 U.S. at 420, 97 S.Ct. at 2775.
On remand, the district court conducted evidentiary hearings November 1 — i, 1977, and in its decision issued December 15,1977, held that:
[TJhere is a burden upon plaintiffs to establish by a preponderance of evidence both a segregative intent and an incremental segregative effect in order to establish a violation of the Equal Protection Clause of the Fourteenth Amendment. (emphasis added). JA-I at 104. Pursuant to this misunderstanding 6 of the Supreme Court’s mandate, the district court individually examined each alleged constitutional violation both for segregative intent and incremental segregative effect. The district court concluded that plaintiffs had failed to meet this burden of proving a constitutional violation and dismissed the complaint. Following the filing of this appeal, this court on January 16,1978, ordered defendants “to cause said system-wide desegregation plan to remain in effect pending appeal, or until further order of this court.”
Appellants and the United States as ami-cus curiae (hereinafter collectively referred to as appellants) contend that various findings of fact and conclusions of law of the district court are both clearly erroneous and are based upon incorrect legal standards. They urge this court to address the legal and factual issues previously reserved in Brinkman I, supra, 503 F.2d 684 and to find that the alleged constitutional violations have a systemwide impact which requires reinstatement of the systemwide remedy approved by this court in Brinkman III, supra, 539 F.2d 1084. Appellants raise four principal assignments of error. First, they contend that the district court misinterpreted the legal relevance of the Board’s conduct prior to the time of Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954) (Brown I), and that the district court’s finding that “[a]t no time did defendant maintain a dual system of education”7 was either based upon the application of incorrect legal standards or was a clearly erroneous factual finding. Appellants argue that as a result of these errors, the district court ignored the principle that if the Board was operating a dual school system at the time of Brown I, or at any time thereafter, it subse*247quently had an affirmative duty to eliminate the systemwide effects of its prior acts of segregation. Second, appellants argue that the district court erred in applying improper legal standards for determining segregative intent. They assert that the district court both failed to utilize the established burden-shifting principles in determining whether various practices were the product of segregative intent and disregarded the established legal standards for determining segregative intent. Third, appellants contend that the district court erred in failing to apply the presumption and burden-shifting principles concerning causation and the impact of unconstitutional conduct. Finally, appellants assert that the district court misalloeated the burden of proof on the issue of the incremental segre-gative effect of the alleged constitutional violations. They argue that the district court erred in holding that plaintiffs were required to demonstrate both the existence of racial discrimination and the specific effects of that discrimination.
Upon a review of the entire record, the arguments of counsel, and upon consideration of the legal and factual issues previously reserved by this court in Brinkman I, supra, 503 F.2d 684, we conclude that the systemwide desegregation plan approved by this court in Brinkman III, supra, 539 F.2d 1084, should be reinstated. The record demonstrates conclusively that at the time of Brown I, defendants intentionally operated a dual school system and that subsequently, defendants never fulfilled their affirmative duty to eliminate the systemwide effects of their prior acts of segregation. To the extent that any findings of fact and conclusions of law of the district court are to the contrary, they are either clearly erroneous, Rule 52, Fed.R.Civ.P., or are incorrect as a matter of law.
I. Pre-Brown violations
This court previously reviewed defendants’ purported intentional segregative acts alleged to have occurred prior to 1954 and concluded that “the Dayton school system has been and is guilty of de jure segregation practices”8 which constituted a “basically dual system,”9 at the time of Brown I. Although we believe this finding to have been implicit in the previous decisions of this court, we now expressly hold that at the time of Brown I, defendants were intentionally operating a dual school system in violation of the Equal Protection Clause of the fourteenth amendment. Our holding is based upon substantial evidence, much of which is undisputed. The finding of the district court to the contrary10 is clearly erroneous, Rule 52, Fed.R.Civ.P., and is based upon both a failure to attribute the proper legal significance to the evidence of pre-Brown I violations and upon various errors of law.
Our review of the record reveals that as of the 1951-52 school year — the last period prior to Brown I for which racial statistics were compiled — the Dayton school board pursued an overt policy of faculty segregation and, through a variety of measures, endeavored to segregate pupils on a racial basis. Defendants admitted that prior to 1951 the Board forbade the assignment of black teachers to white or mixed classrooms “pursuant to an explicit segregation policy.” 11 The district court found that “until *2481951 the Board’s policy of hiring and assigning faculty was purposefully segrega-tive.” 12 A review of the record establishes to our satisfaction that the assignment of faculty was purposefully segregative;13 but contrary to the finding of the district court, we found in Brinkman I, supra, 503 F.2d at 697 — 98 that the Board “effectively continued in practice the racial assignment of faculty through the 1970-71 school year.” To the extent that the finding of the district court is contrary to the conclusion of this court, it is clearly erroneous.
The undisputed evidence reflects that during the 1951-52 school year, the faculty at the four 100 percent black schools (Garfield, Dunbar, Willard, and Wogamon) was 100 percent black whereas with one exception,14 the faculty at all other schools in the system was 100 percent white.15 Defendants further admitted that as of 1954, 91.4 percent of the 162 non-travelling black teachers were assigned to schools with all black student populations.16 Thus, at the time of Brown I, it was possible to identify a “black school” in the Dayton system without reference to the racial composition of pupils. In Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 18, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971), the Supreme Court stated:
Independent of student assignment, where it is possible to identify a ‘white school’ or a ‘Negro school’ simply by reference to the racial composition of teachers and staff ... a prima facie case of violation of substantive constitutional rights under the Equal Protection Clause is shown.17
The district court, however, failed to attribute the proper legal significance to the deliberate policy of faculty segregation adopted and applied by defendants.
The purposeful segregation of faculty by race was inextricably tied to racially motivated student assignment practices. The record reflects that in the 1951 — 52 school year, 77.6 percent of all students attended schools in which one race accounted for 90 percent or more of the students and 54.3 percent of the black students were assigned *249to the four schools that were 100 percent black.18 We recognize that racial imbalance in student attendance patterns is not in itself a constitutional violation. See Dayton Board of Education v. Brinkman, supra, 433 U.S. at 413, 417, 97 S.Ct. 2766, 53 L.Ed.2d 851 (1977); Washington v. Davis, 426 U.S. 229, 240, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976); Keyes v. School District No. 1, 413 U.S. 189, 198, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973). However, such racial imbalance does assume increased significance in the historical context of repeated intentional segregative acts by the school board directed at the four schools.which were 100 percent black in 1954. See Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 267, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977). Defendants contend that such evidence of pre-Brown I constitutional violations is irrelevant, or, alternatively, that the effects of any past intentional segregative actions have become attenuated in the ensuing years. These contentions are wholly without merit. First, with respect to evidence of pre-Brown I constitutional violations, the Supreme Court noted in Keyes, supra, 413 U.S. at 210-11, 93 S.Ct. at 2698 that:
We reject any suggestion that remoteness in time has any relevance to the issue of intent. If the actions of school authorities were to any degree motivated by segregative intent and the segregation resulting from those actions continues to exist, the fact of remoteness in time certainly does not make those actions any less ‘intentional.’
Second, with respect to the question of attenuation, defendants have failed to meet their burden of proving that the effects of any past intentional actions have become attenuated. Keyes, supra, 413 U.S. at 211, 93 S.Ct. 2686.
Garfield school was the site of intra-school racial segregation which began in 1912 and was ruled illegal by the Supreme Court of Ohio in Board of Education of School District of City of Dayton v. State ex rel. Reese, 114 Ohio St. 188, 151 N.E. 39 (1926).19 Defendant Wayne Carle, Superintendent of the Dayton schools, admitted, however, that racial segregation continued virtually unabated at Garfield after the Reese decision,20 and that during the 1930’s, white students who lived in the Garfield attendance area were permitted to transfer to predominantly white schools.21 As a result of the actions of the Board, Garfield became all black in student enrollment in 1936 and, at approximately the same time, an all black faculty was assigned to the school.22 Thereafter, Garfield was maintained as an all black school.
The district court found that Dunbar high school intentionally had been established as a district-wide school for only black students with an all black faculty and a black principal.23 The record reveals that black students throughout Dayton automatically were assigned or otherwise were induced to attend Dunbar and that, in many instances, black students crossed attendance boundaries to do so.24 Defendants further admitted that until approximately 1947, Dunbar was not allowed to participate in the city athletic conference and consequent*250ly, Dunbar athletic teams played other all black high schools from other cities.25 Defendants also admitted that until several months after the decision in Brown I, black children were transported by bus from an orphanage past white schools to Dunbar.26 The district court found that this practice was “arguably ... a purposeful seg-regative act.”27 To the extent that this finding implies that this practice was not purposefully segregative, it is clearly erroneous.
Defendants assert that since attendance at Dunbar was voluntary, there is no justification for finding that the establishment and operation of the school constituted intentionally segregative acts. This argument misses the point entirely. First, until at least as late as 1952, the option of attending Dunbar was available only to blacks since, pursuant to school board policy, whites could not be taught by Dunbar’s all black faculty. Second, the record reflects that many black children were automatically assigned or otherwise encouraged to attend Dunbar regardless of choice.28 Finally, the record indicates that the “choice” of attending Dunbar, in many instances, may have been merely a less drastic alternative than attending other schools which practiced intra-school segregation and discrimination.29
In this manner and through these procedures, the Board intentionally operated Dunbar as an all black school until it was closed as a high school in 1962. The operation of Dunbar clearly had the effect of keeping other high schools throughout the district predominantly white during those years.30
The record reflects that during the early 1940’s, the student body of Wogamon elementary school became predominantly black in part because the Board permitted white students to transfer to predominantly white schools.31 In June 1945, Wogamon closed with an all white staff and reopened in September 1945 with an all black staff and a black principal.32 Wogamon subsequently became and presently is an all black school. Similarly, the record reflects that in the 1930’s the Willard school became predominantly black due to increased black enrollment and the transfer of white students. The record indicates that in 1934, Willard school had a 50 percent black student body and a faculty which was 38 percent black. The following year, however, the student body became approximately 95 *251percent black with an all black faculty.33 By 1947, Willard was 100 percent black in student enrollment and subsequently it has remained a one race school.
Additional evidence also establishes that prior to 1954, the Board pursued a policy of racial separation. Defendants admit that until approximately 1950, “separate facilities, including separate swimming pools and locker room facilities were maintained at Roosevelt [school] for black and white students.” 34 In addition, during the late 1940’s and early 1950’s, defendants operated one race classrooms in officially one race housing projects which the district court found were “strictly segregated according to race.”35
Upon a review of this evidence, the relevant inquiry is whether at the time of Brown I, or any time thereafter, defendants were operating a dual school system in violation of the Equal Protection Clause of the fourteenth amendment. In Keyes v. School District No. 1, supra, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548, the Supreme Court held that in order to establish a violation of the fourteenth amendment in school desegregation cases where no statutory dual system has ever existed, plaintiffs must demonstrate purposeful state imposed segregation in a substantial portion of the school system.36
In Brinkman II, supra, 518 F.2d at 854, this court held that defendants had been guilty of de jure segregative practices. There is ample evidence to support the finding that at the time of Brown I defendants were carrying out “a systematic program of segregation affecting a substantial portion of the students, schools, teachers, and facilities.” 37 As noted previously, at the time of Brown I, approximately 54.3 percent of the black pupils in the Dayton school system were assigned to four schools that had all black faculties and student bodies. In Keyes, supra, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548, the finding that the Denver school board was guilty of intentional seg-regative acts with respect to schools attended by only 37.69 percent of Denver’s black students was sufficient to constitute the entire school district a dual system. The finding of the district court that defendants never had operated a dual school system38 is clearly erroneous and is based upon misconceptions of the applicable law.
The district court erred both in failing to accord the proper legal significance to the facts extant at the time of Brown I and in failing to apply the appropriate presumption and burden-shifting principles of law. The district court failed to attribute the proper legal significance to the deliberate policy of faculty segregation which, at the time of Brown I, made it possible to identify a “black school” in the Dayton system without reference to the racial composition of pupils.39 The district court also failed to attribute the proper legal significance to the evidence that at the time of Brown I, Garfield, Willard, Wogamon and Dunbar schools were deliberately segregated or racially imbalanced due to the actions of defendants. These facts were sufficient to constitute a prima facie violation of the fourteenth amendment under the rule of Swann, supra, 402 U.S. at 18, 91 S.Ct. 126740 and to shift the burden of proof to defendants. The district court also misconstrued the proper approach for determining discriminatory purpose and intent which *252may be inferred from objective circumstantial evidence41 and through the use of reasonable presumptions.42 This court stated in Oliver v. Michigan State Board of Education, 508 F.2d 178, 182 (6th Cir. 1974), cert. denied, 421 U.S. 963, 95 S.Ct. 1950, 44 L.Ed.2d 449 (1975) that:
A presumption of segregative purpose arises when plaintiffs establish that the natural, probable, and foreseeable result of public officials’ action or inaction was an increase or perpetuation of public school segregation. The presumption becomes proof unless defendants affirmatively establish that their action or inaction was a consistent and resolute application of racially neutral policies, (citations omitted).
Accord, Arthur v. Nyquist, 573 F.2d 134 (2d Cir.. 1978); NAACP v. Lansing Board of Education, 559 F.2d 1042, 1047-48 (6th Cir.), cert. denied, 434 U.S. 997, 98 S.Ct. 635, 54 L.Ed.2d 491 (1977); Bronson v. Board of Education, 525 F.2d 344 (6th Cir. 1975), cert. denied, 425 U.S. 934, 96 S.Ct. 1665, 48 L.Ed.2d 175 (1976); Hart v. Community School Board of Education, 512 F.2d 37 (2d Cir. 1975). The evidence clearly establishes that the natural, probable and foreseeable result of defendants’ actions was the creation and perpetuation of a dual school system. The district court, moreover, failed to recognize the teaching of Keyes, supra, 413 U.S. at 208, 93 S.Ct. at 2697, that:
[A] finding of intentionally segregative school board actions in a meaningful portion of a school system, as in this case, creates a presumption that other segregated schooling within the system is not adventitious. It establishes, in other words, a prima facie case of unlawful segregative design on the part of school authorities, and shifts to those authorities the burden of proving that other segregated schools within the system are not also the result of intentionally segrega-tive actions. This is true even if it is determined that different areas of the school district should be viewed independently of each other because, even in that situation, there is high probability that where school authorities have effectuated an intentionally segregative policy in a meaningful portion of the school system, similar impermissible considerations have motivated their actions in other areas of the system.
The district court erred in failing to shift the burden of proof to defendants.
A review of the entire record indicates that defendants have not established that the character of the school system extant in 1954 was the result of racially neutral acts. We emphasize that defendants’ intentional segregative practices cannot be confined in one distinct area.43 To the contrary, defendants’ segregative practices at the time of Brown I infected the entire Dayton public school system. There is no doubt that “racially inspired school board actions have an impact beyond the particular schools that are the subjects of those actions.” Keyes, supra, 413 U.S. at 203, 93 S.Ct. at 2695, and that the effect of the operation of this dual school system was to maintain other schools in the district as predominantly white.44
*253II. Post-Brown violations
The district court’s error in failing to find that defendants were operating a dual school system at the time of Brown I resulted also in its failure to evaluate properly the Board’s post-Brown I actions, which must be judged by their efficacy in eliminating the continuing effects of past discrimination. In Brinkman I, supra, 503 F.2d at 704, this court stated:
Once the plaintiffs-appellants have shown that state-imposed segregation existed at the time of Brown (or any point thereafter), school authorities 'automatically assume an affirmative duty . . . to eliminate from the public schools within their school system “all vestiges of state-imposed school segregation.” ’ Keyes, supra, 413 U.S. at 200, 93 S.Ct. at 2693.
Thus, for 24 years defendants have been under a constitutional duty to desegregate the Dayton public schools. See Penick v. Columbus Board of Education, 583 F.2d 787, at 799 (6th Cir. 1978). The district court specifically found that “with one exception ... no attempt was made to alter the racial characteristics of any of the schools” and that the one exception “was in fact a failure.”45 The district court, however, neither charged defendants with the affirmative duty to eliminate the effects of their discrimination nor did it place upon the Board the burden of proving that it had done so. The evidence of record demonstrates convincingly that defendants have failed to eliminate the continuing system-wide effects of their prior discrimination and have intentionally maintained a segregated school system down to the time the complaint was filed in the present case. In addition, the record discloses post-1954 actions which actually have exacerbated the racial separation existing at the time of Brown I.
A. Faculty and student assignment practices
In Brinkman I, supra, 503 F.2d at 697 — 98, this court found that defendants “effectively continued in practice the racial assignment of faculty through the 1970-71 school year.”46 This finding is supported by substantial evidence on the record.47 The finding of the district court to the contrary 48 is clearly erroneous. Rule 52, Fed.R.Civ.P. The district court also erred in failing to attribute the correct legal significance to the persistently discriminatory faculty assignment practices as a component of the Board’s perpetuation of the dual system extant at the time of Brown I. Moreover, the district court again failed to recognize this proof of continuing purposeful segrega-tive acts as an element of plaintiffs’ prima facie case.49 The effect of having established this prima5 facie case should have been to shift to the Board the burden of rebutting the presumption that other practices likewise were undertaken with segre-gative intent.
For example, in 1962 the Willard and Garfield schools, previously operated for blacks only, were closed and the all black Dunbar high school building was converted into McFarlane elementary school. Most of the children from the Willard and Garfield attendance areas simply were assigned to the McFarlane school which opened with an all black student body and an all black faculty. Some children from the Willard and Garfield areas also were assigned to the all black Miami Chapel and Irving elementary schools. Simultaneously, the new Dunbar high school opened with a virtually *254all black student body and faculty. Defendants should have been required to rebut the reasonable presumption that the simultaneous assignment of both a predominantly black faculty and student body at these schools was the product of segrega-tive intent and an effort to perpetuate the dual school system extant at the time of Brown I.
This error was compounded by imposing upon plaintiffs the additional burden of proving specific causal relationships between the widespread faculty segregation practices and the substantial student segregation existing at the time of trial.
Nowhere in the record do defendants convincingly demonstrate that the systemwide student racial imbalance characteristic of the Dayton public school system since at least the time of Brown I likewise was not the product of segregative acts. Keyes, supra, 413 U.S. at 210, 93 S.Ct. 2686, 2698. “[I]t is not enough . . . that school authorities rely upon some allegedly logical, racially neutral explanation.” Id. Defendants here have failed “to adduce proof sufficient to support a finding that segregative intent was not among the factors that motivated their actions.” Id. The Court in Keyes noted further that:
[I]f respondent School Board cannot disprove segregative intent, it can rebut the prima facie case only by showing that its past segregative acts did not create or contribute to the current segregated condition of the core city schools. Id. at 211, 93 S.Ct. at 2699.
Defendants have failed to establish that their prior segregative acts did not create or contribute to the current segregated condition of the Dayton schools.
In Brinkman I, supra, 503 F.2d at 694-95, this court stated that:
Enrollment data from the Dayton system reveals the substantial lack of progress that has been made over the past 23 years in integrating the Dayton school system. In 1951-52, of 47 schools, 38 had student enrollments 90 per cent or more one race (4 black, 34 white). Of the 35,000 pupils in the district, 19 per cent were black. Yet over half of all black pupils were enrolled in the four all black schools; and 77.6 per cent of all pupils were assigned to virtual one race schools. ‘Virtual one race schools’ refers to schools with student enrollments of 90 per cent or more one race. In 1963-64, of 64 schools, 57 had student enrollments 90 per cent or more one race (13 black, 44 white). Of the 57,400 pupils in the district, 27.8 per cent were black. Yet 79.2 per cent of all black pupils were enrolled in the 13 black schools; and 88.8 per cent of all pupils were enrolled in such one race schools.
In 1971-72 (the year the complaint was filed), of 69 schools, 49 had student enrollments 90 per cent or more one race (21 black, 28 white). Of the 54,000 pupils 42.7 per cent were black; and 75.9 per cent of all black students were assigned to the 21 black schools. In 1972-73 (the year the hearing was held) of 68 schools, 47 were virtually one race (22 black, 25 white); fully 80 per cent of all classrooms were virtually one race. (Of the 50,000 pupils in the district, 44.6 per cent were black).
Every school which was 90 per cent or more black in 1951-52 or 1963-64 or 1971-72 and which is still in use today remains 90 per cent or more black. Of the 25 white schools in 1972-73, all opened 90 per cent or more white and, if open, were 90 per cent or more white in 1971-72, 1963-64 and 1951-52.
Nowhere in the record have defendants demonstrated that the present systemwide racial imbalance would have occurred even in the absence of their segregative acts. As the Supreme Court noted in Swann, supra, 402 U.S. at 26, 91 S.Ct. at 1281, there is a presumption against schools that are “substantially disproportionate in their racial composition” in school systems with a history of segregation, as in Dayton.50
*255The conclusion that the maintenance of persistent racial imbalance in the Dayton schools was not merely adventitious is bolstered by defendants’ use of optional attendance zones for racially discriminatory purposes in clear violation of the Equal Protection Clause.51 In 1973, the district court determined that some optional attendance zones had been created intentionally for racially segregative purposes and that the zones had demonstrable racial effects.52 These findings of fact were affirmed by this court in Brinkman I, supra, 503 F.2d at 696, and are supported by substantial evidence. Nevertheless, following remand from the Supreme Court, the district court repudiated these findings, concluding that “[n]o evidence has been presented suggesting that attendance zones were redrawn to promote segregation”53 and that the zones had no segregative effect.54 In reaching these clearly erroneous findings of fact, the district court once again failed to recognize the optional zones as a perpetuation, rather than an elimination, of the existing dual system; failed to afford plaintiffs the burden-shifting benefits of their prima facie case; and failed to evaluate the evidence in light of tests for segregative intent enunciated by the Supreme Court, this court and other circuits in decisions cited in this opinion.
B. School construction and site selection
The evidence of record establishes that of 24 new schools constructed between 1950 and the time this action was instituted, 22 opened 90 percent or more black or white.55 During the same period, 78 of the 86 additions of classroom space for which racial compositions are known were made to schools 90 percent or more one race.56 Coupled with these practices were some instances of the coordinate racial assignment of professional staffs to these schools and additions on the basis of the racial composition of the pupils served by the schools.57 This court noted in NAACP v. Lansing Board of Education, 559 F.2d 1042, 1056 (6th Cir.), cert. denied, 434 U.S. 997, 98 S.Ct. 635, 54 L.Ed.2d 491 (1977) that “[s]chool construction which promotes racial imbalance or isolation is an important indicium of a de jure segregated school system.” See Oliver v. Michigan State Board of Education, supra, 508 F.2d at 184. See generally United States v. School District of Omaha, 521 F.2d 530, 543-46 (8th Cir. 1975), cert. denied, 423 U.S. 946, 96 S.Ct. 361, 46 L.Ed.2d 280 (1976). In the face of this, the district court failed to infer purposeful segregation from this pattern of school construction which unmistakably increased or maintained racial isolation.58 Again the district court failed to recognize that plaintiffs had established a prima facie constitutional violation which shifted the burden of proof to defendants. Instead, the district court concluded that plaintiffs had failed to show that defendants’ site selection and construction practices “had a segregative purpose or . had an incremental segregative effect upon pupils, teachers, or staff.” 59 These findings of fact are infected by legal error and are clearly erroneous. As detailed previously, the post-Brown I *256practices of racially motivated faculty assignments to new schools bespeaks a concomitant segregative intent in the location of new schools and additions. Nowhere in the record have defendants established that their school construction and site selection practices and the simultaneous racially motivated assignment of teachers were the product of racially neutral policies. Defendants have failed “to adduce proof sufficient to support a finding that segregative intent was not among the factors that motivated their actions.” Keyes supra, 413 U.S. at 210, 93 S.Ct. at 2698.
The district court’s conclusion that defendants’ school construction and site selection practices had no segregative effect likewise is clearly erroneous. Instead of meeting their affirmative duty to disestablish the dual school system extant at the time of Brown I and to diffuse black and white students throughout the Dayton school system, defendants pursued a policy of containment through school construction and site selection practices. As noted previously, at the time of the initial hearings in this case, approximately 80 percent of all classrooms in the Dayton school system were virtually one race. On the basis of the evidence of record, the conclusion is inescapable that defendants’ school construction and site selection practices were segregative in effect.
C. Grade structure and reorganization
Appellants’ principal objection in this area is to the establishment in the 1971-72 school year of a middle school system which allegedly had a segregative effect. In a report issued in 1971, the Ohio Department of Education characterized the middle school system as the apparent addition of
one more action to a long list of state-imposed activities which are offensive to the Constitution and which are degrading to school children. Along with many other affirmative duties which the Dayton Board must fulfill, correction of this particular offense must occur. PX 12, JA-V at 454.
The report further opined that:
Of the five sets of schools currently involved in the process of conversion to feeder and middle schools, the following seems to be occurring:
1. two sets of schools will be totally black;
2. racial isolation will actually be increased in one set of schools; and
3. only in the Dayton View area, which was previously integrated, could conversion to middle schools possibly result in reduction of racial and economic isolation and insulation. Id.
Unrebutted testimony concluded that the effect of the middle school system was to increase or maintain segregation rather than to eradicate it in accordance with defendants’ affirmative duty to disestablish the dual system.60 The district court found that the middle schools had both “a segre-gative effect and an integrative effect."61 Nevertheless, the district court concluded that plaintiffs had failed to establish segre-gative intent in the establishment of the middle schools. This finding is questionable in light of plaintiffs’ convincing demonstration that the natural, probable, and foreseeable result of the establishment of the middle schools was an increase or perpetuation of segregation. The district court failed to recognize the middle school system as one of the areas in which defendants failed to disestablish Dayton’s dual school system.
Upon consideration of the record, the conclusion is inescapable that, rather than eradicate the systemwide effects of the dual system extant at the time of Brown I, defendants’ racially motivated policies with respect to the assignment of faculty and students, use of optional attendance zones, school construction and site selection, and grade structure and reorganization perpetuated or increased public school segregation in Dayton. Thus, defendants have utterly failed to comply with their ongoing 24 year *257obligation to desegregate the Dayton public schools, Penick v. Columbus Board of Education, supra, 583 F.2d at 799, and, in addition, have committed affirmative acts that have exacerbated the existing racial segregation. The remedy directed in this opinion is made necessary by: (1) the failure of defendants to disestablish the pre-1954 segregated school system; and (2) post-1954 acts of systemwide impact which have contributed affirmatively to the continuation of a segregated system.
III. Remedy
In Dayton Board of Education v. Brinkman, supra, 433 U.S. at 420, 97 S.Ct. at 2775, the Supreme Court stated that upon finding a constitutional violation:
[T]he District Court in the first instance, subject to review by the Court of Appeals, must determine how much incremental segregative effect, these violations had on the racial distribution of the Dayton school population as presently constituted, when that distribution is compared to what it would have been in the absence of such constitutional violations. The remedy must be designed to redress that difference, and only if there has been a systemwide impact may there be a systemwide remedy. Keyes, 413 U.S. at 213, 93 S.Ct. 2686. (emphasis added).
Contrary to the conclusion of the district court,62 we are convinced that the term “incremental segregative effect” used by the Supreme Court in the Brinkman decision, was not intended to change the standards for fashioning remedies in school desegregation cases. Penick v. Columbus Board of Education, supra, 583 F.2d at 787, 794, 813; NAACP v. Lansing Board of Education, 581 F.2d 115, (6th Cir. 1978), cert. denied, -U.S.-, 98 S.Ct. 3126, 57 L.Ed.2d 1150 (1978). The purpose of the remedy is to eliminate the lingering effects of intentional constitutional violations and to restore plaintiffs to substantially the position they would have occupied in the absence of these violations. The word “incremental” merely describes the manner in which segregative impact occurs in a northern school case where each act, even if minor in itself, adds incrementally to the ultimate condition of segregated schools. The impact is “incremental” in that it occurs gradually over the years instead of all at once as in a case where segregation was mandated by state statute or a provision of a state constitution.
The district court committed two errors in its approach to this inquiry. First, it individually examined each alleged constitutional violation as if it were an isolated occurrence and sought to determine the incremental segregative effect of that occurrence. In Keyes, supra, 413 U.S. at 200, 93 S.Ct. at 2693, the Court stated:
We have never suggested that plaintiffs in school desegregation cases must bear the burden of proving the elements of de jure segregation as to each and every school or each and every student within the school system. Rather, we have held that where plaintiffs prove that a current condition of segregated schooling exists within a school district where a dual system was compelled or authorized by statute at the time of our decision in Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954) (Brown I), the State automatically assumes an affirmative duty ‘to effectuate a transition to a racially nondiscriminatory school system,’ Brown v. Board of Education, 349 U.S. 294, 301, 75 S.Ct. 753, 99 L.Ed. 1083 (1955) (Brown II), see also Green v. County School Board, 391 U.S. 430, 437-438, 88 S.Ct. 1689, 20 L.Ed.2d 716 (1968), that is, to eliminate from the public schools within their school system ‘all vestiges of state-imposed segregation.’ Swan v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 15, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971).
The district court’s act-by-act approach is no more valid than the school-by-school approach rejected in Keyes. As this court noted in Penick, supra, 583 F.2d at 814:
Dayton does not . . . require each of fifty segregative practices or episodes to be judged solely upon its sepa*258rate impact on the system. The question posed concerns the impact of the total amount of segregation found — after each separate practice or episode had added its ‘increment’ to the whole. It was not just the last wave which breached the dike and caused the flood.
Secondly, the district court erred in allocating the burden of proof on the issue of incremental segregative effect to plaintiffs, requiring them to establish both racial discrimination and the specific incremental effect of that discrimination. Where plaintiffs prove, as here, a systemwide pattern of intentionally segregative actions by defendants, it is the defendants’ burden to overcome the presumption that the current racial composition of the school population reflects the systemwide impact of those violations. See Keyes, supra, 413 U.S. at 211 n. 17, 93 S.Ct. 2686. Nowhere in the record have defendants rebutted this presumption. Since the district court failed to apply the proper legal standards, we independently consider the incremental segregative effect of defendants’ most egregious practices. In so doing, we are mindful that “racially inspired school board actions have an impact beyond the particular schools that are the subjects of those actions.” Keyes, supra, 413 U.S. at 203, 93 S.Ct. at 2695. First, the dual school system extant at the time of Brown I embraced “a systemwide program of segregation affecting a substantial portion of the schools, teachers, and facilities”63 of the Dayton schools, and, thus, clearly had systemwide impact. See Penick v. Columbus Board of Education, supra, 583 F.2d at 814, 815. Secondly, the post-1954 failure of defendants to desegregate the school system in contravention of their affirmative constitutional duty obviously had systemwide impact. Id. at 815. The impact of defendants’ practices with respect to the assignment of faculty and students, use of optional attendance zones, school construction and site selection, and grade structure and reorganization clearly was systemwide in that the actions perpetuated and increased public school segregation in Dayton.
We hold further that each of defendants’ policies and practices detailed in this opinion added an increment to the sum total of the constitutional violations.
Finding that the constitutional violations before the court have a systemwide impact, Brinkman, supra, 433 U.S. at 420, 97 S.Ct. 2766, we conclude that the system-wide desegregation plan approved by this court in Brinkman III, supra, 539 F.2d 1084, should be reinstated. This remedy is “tailored to undo the violations of plaintiffs’ constitutional rights . . .” and is “designed to redress” the effect of the violations found. NAACP v. Lansing Board of Education, 581 F.2d 115, supra (6th Cir. 1978), cert. denied, -U.S.-, 98 S.Ct. 3126, 57 L.Ed.2d 1150 (1978). The decision of the district court is reversed. It is ordered that the desegregation plan approved by this court in Brinkman III, supra, 539 F.2d 1084, be and hereby is reinstated and shall remain in effect during the 1978 — 79 school year. Plaintiffs-appellants shall recover the costs of this appeal from the Dayton Board of Education. The case is remanded to the district court for further proceedings not inconsistent with this opinion.