OPINION
This is a school desegregation case brought by black parents and their chil*415dren who attend the Boston public schools. Plaintiffs seek for themselves and on behalf of their class1 declaratory and injunctive relief against the defendants for a myriad of acts that allegedly violate the constitutional rights of the plaintiff class. Defendants are the Boston School Committee, its individual members, and the Superintendent of the Boston Public Schools (hereinafter collectively “the city defendants”), and the Board of Education of the Commonwealth of Massachusetts, its individual members, and the Commissioner of Education (hereinafter collectively “the state defendants”).
Plaintiffs have alleged that the city defendants have intentionally brought about and maintained racial segregation in the Boston public schools by various actions, including the adoption and maintenance of pupil assignment policies, the establishment and manipulation of attendance areas and district lines reflecting segregated residential patterns, the establishment of grade structures and feeder patterns, the administration of school capacity, enlargement, and construction policies, transportaion practices, and by unjustifiably failing to adopt or implement policies reasonably available to eliminate racial segregation in the Boston public schools. Plaintiffs assert that these alleged practices have resulted in denying black school children the equal protection of the laws, in violation of the Fourteenth Amendment to the United States Constitution. See Brown v. Board of Education, 1954, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873; Keyes v. School Dist. No. 1, 1973, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548. Plaintiffs further contend that the city defendants and their predecessors have engaged in racial discrimination with respect to the hiring and assignment of faculty and staff, and with respect to curricula and the allocation of instructional materials, and resources; that both the city and state defendants have implemented pupil classification practices which discriminate against some children in admission to certain schools, and have maintained a pattern of lower instructional expenditures in schools attended disproportionately by black children. Plaintiffs argue that these practices deny black children their constitutional right to equality of educational opportunity under the Fourteenth Amendment. Federal jurisdiction is invoked and exists under 28 U.S.C. § 1343, and violations of the Thirteenth and Fourteenth Amendments and 42 U.S.C. §§ 1981,1983 and 2000d are alleged.
The city defendants have generally denied the allegations of the plaintiffs. They have also argued that to the extent that schools in the Boston system contain disproportionate numbers of whites or blacks, that result is due to residential segregation over which they have no control and also due to the neighborhood school policy, which defendants claim is a constitutionally permissible tool of educational policy. The city defendants further assert that not only have they not promoted segregation but they have tried to effect a better racial balance in the Boston public schools in accordance with the Racial Imbalance Act, Mass.G. L. c. 71, §§ 37C and 37D, and they allege that they have been found to be in compliance with that Act. Along this same *416line of defense, the city defendants assert that they can do no more consistently with the Racial Imbalance Act than they have already done to eliminate racial imbalance in the city’s public schools.
The state defendants have also denied any constitutional wrongdoing. They contend that they have frequently made unheeded suggestions to the. city defendants on how to alleviate racial imbalance in Boston; that they have only limited control over the activities of the city defendants ; and that by enforcing the Racial Imbalance Act and by pursuing judicial enforcement of the Act when the board of education’s orders were not complied with by the Boston School Committee they have done as much as possible to eliminate racial segregation in the Boston public schools. The state defendants have agreed with virtually all the contentions made by the plaintiffs here against the city defendants.
Various pretrial rulings should be mentioned. Shortly after the action was commenced, the members and officers of the Boston Teachers Union sought to intervene as parties defendant. Partly because the positions of the city defendants and the teachers were identical on constitutional issues involved in this case, the court denied the motion without prejudice to renewal should the alleged constitutional violations be found. The court denied a motion of the city defendants to join numerous cities and towns around Boston as defendants, partly on the ground that the proposed defendant cities and towns had not been charged by the plaintiffs with contributing to the violation of their constitutional rights.2 3 The court also denied a motion to dismiss the complaint by the city defendants, and their motion to stay the proceedings pending implementation of an order entered against the city defendants in a state court case between it and the state defendants involving the Massachusetts Racial Imbalance Act, Mass. G.L. c. 71, §§ 37C, 37D. School Committee of Boston v. Board of Education, 1973 Mass.Adv.Sh. 161, 292 N.E.2d 338. The court granted a motion to dismiss a crossclaim of the state defendants charging the city defendants with actions substantially set forth in plaintiffs’ own complaint.
At the court’s request, the parties have attempted, insofar as possible, to enter into stipulations of undisputed facts and to introduce evidence by way of designations and counter-designations of depositions and prior testimony of witnesses in the state case referred to above and in a proceeding between the city and the federal Department of Health, Education & Welfare. Counsel’s pretrial activities helped to shorten the actual trial of this case, which lasted fifteen days, and which concerned only the liability issues of the case, as contrasted with issues relating to the possible remedy. The court also had a one-half day view of some of the Boston schools and received several hundred exhibits.
The case was reopened on June 20, 1973 to receive evidence regarding the use of a new high school facility. The plaintiffs alleged that the city defendants in their conduct with respect to this new structure were engaged in further racial discrimination in violation of the Equal Protection Clause. The court received further evidence from both parties. On March 15, 1974 the city defendants moved to reopen the case to submit evidence dealing with their ongoing litigation in the state courts *417with the State Board of Education and the plan they had been ordered to implement under the Racial Imbalance Act. This motion was denied.
Massachusetts Racial Imbalance Act
Questions of racial segregation in the public schools of Massachusetts have a unique dimension because of a state racial imbalance law, Mass.G.L. c. 71, §§ 37C, and 37D and c. 15, §§ II, 1J and IK, passed by the legislature in 1965. It has been said that the requirements of the state statute go further than those of the Fourteenth Amendment, see School Committee of Springfield v. Board of Education, 1974 Mass.Adv.Sh. 657, 673, 311 N.E.2d 89 (May 1, 1974— Springfield II), in the sense that the statute commands affirmative action to eliminate racial imbalance in public school systems whatever its cause, i. e., independently of a finding of de jure segregation. The statute has been construed and applied in several decisions of the Supreme Judicial Court of the Commonwealth pertaining to public schools in Boston and Springfield. See, as to the latter, School Committee of Springfield v. Board of Education, 1973 Mass.Adv.Sh. 1543, 287 N.E.2d 438, and Springfield II, supra. Although the statute is not directly involved in the instant proceedings, it is highly relevant. Many of the defendants’ actions were taken as a result of the state law. The provisions of the statute were the subject of frequent discussion at meetings of the Boston School Committee. Efforts by the city defendants to evade the statute illumine their intent with respect to school segregation generally. Communications between the city and state defendants were usually in the context of provisions of the statute.
The Racial Imbalance Act evolved from a report in April 1965 by an advisory committee appointed by the State Board of Education and the Commissioner of Education to study racial segregation in the public schools of Massachusetts generally and of Boston, Springfield, Cambridge, Medford and Worcester individually, named the Kiernan Report after the then Commissioner of Education for the Commonwealth, Dr. Owen B. Kiernan. The report concluded that racial imbalance was educationally harmful and should be eliminated. The report specifically noted that Boston contained 45 “imbalanced” schools — i. e., schools with more than 50% non-white students, and proposed various methods whereby the city might solve the problem. The response of the Boston School Committee was first, on April 26, 1965, to refuse to acknowledge that racial imbalance was a problem that ought to be eliminated and then in June to adopt a report by Superintendent Ohrenberger which disputed the premise that racial imbalance was educationally harmful but suggested that “interracial learning experiences are socially desirable.” As a result of the Kiernan Report, the Massachusetts legislature passed the Racial Imbalance Act of 1965.3 The statute declares that the promotion of racial balance in public schools is the policy of the Commonwealth and provides that “[t]he prevention or elimination of racial imbalance shall be an objective in all decisions involving the drawing or altering of school attendance lines and the selection of new school sites.” Mass.G.L. c. 71, § 37C. Under the Act, the school committees in Massachusetts cities and towns must each year file racial statistics with the State Board of Education. If the state board finds “racial imbalance”, which exists when “the percent' of *418non-white students in any public school is in excess of fifty percent of the total number of students in such school”, it notifies the school committee of that fact. The committee must then file a plan for the elimination of such racial imbalance. Plans for eliminating imbalance may involve redistricting, new school construction, additions to schools and “other methods.” Under the statute only safety may be considered equally with the alleviation of racial imbalance. Regarding the sensitive question of busing, the Act provides,
“No school committee or regional school district committee shall be required as part of its plan to transport any pupil to any school outside its jurisdiction or to any school outside the school district established for his neighborhood, if the parent or guardian of such pupil files written objection thereto with such school committee.”
The statute does not mention racial imbalance of faculty and administrative staff.
Under the Act it is the responsibility of the state board to provide assistance in the formulation of plans; to make recommendations if no plan is filed, or if a plan is filed but does not meet statutory requirements; and to withhold state financial aid if the school committee “does not show progress within a reasonable time in eliminating racial imbalance in its schools . . . . ” Mass. G.L. c. 15, § II. With respect to state financial assistance, the usual state contribution in Massachusetts to the cost of schoolhouse construction or enlargement is 40%; the Act provides, in § II, for an increase to 65% when construction is for the purpose of reducing or eliminating racial imbalance in the school system.
State Court Litigation
State courts have made several rulings whether the Boston school committee has complied with the Racial Imbalance Act and whether the state board might lawfully withhold state aid from the City of Boston. More often than not their rulings have turned on procedural aspects of the Racial Imbalance Act. Nevertheless a general familiarity with the state court litigation is essential to a full understanding of the Boston situation. The first suit by the committee was the attack upon the constitutionality of the Racial Imbalance Act in 1967, noted ante at n. 3. In a subsequent case brought by the committee in the Suffolk County Superior Court, the withholding of state funds by the board under the Act was challenged. The controversy centered around what was known as the city’s Fourth Stage Racial Imbalance Plan.4 On December 18, 1972 the Superior Court ruled, in a complex opinion, that the board could not withhold the state monies because the committee was in literal compliance with the Act, noting, however, that compliance was minimal. On appeal the Massachusetts Supreme Judicial Court affirmed the Superior Court decision on the ground that the board’s action had been predicated upon a single committee vote on September 21, 1971 5 rather than upon all of the circumstances of imbalance within Boston. School Committee of Boston v. Board of Education, 1973 Mass.Adv.Sh. 161, 292 N.E.2d 338 (February 2, 1973). On February 15, 1973 the Supreme Judicial Court reviewed a determination of invalidity of a board plan which arose *419out of the same Superior Court proceedings. The court ruled that judicial review of the board’s plan should be upon the basis of an administrative record and ordered the trial judge to vacate his order and interlocutory decree and to remand the proceedings to the board for the development of a record. Professor Jaffe of Harvard Law School, as a hearing examiner for the board, conducted hearings and filed a detailed report on May 29, 1973, including findings and recommendations which were substantially adopted by the board on June 25, 1973.
On June 28, 1973 Governor Sargent requested an advisory opinion from the Supreme Judicial Court on the constitutionality of a bill passed by the legislature and awaiting his approval which would prohibit the transportation of students without the written consent of their parents. In Opinion of the Justices, 1973 Mass.Adv.Sh. 1027, 298 N.E. 2d 840 (July 10, 1973), the court ruled that the bill would be unconstitutional and the Governor vetoed it.
On July 16, 1973 the Supreme Judicial Court reviewed three suits involving the use of a new $24,000,000, ten-story building on Avenue Louis Pasteur, due to open in September 1973. This tower facility was constructed as part of a plan to alleviate racial imbalance in high schools and accordingly qualified for increased state financial assistance, 65% of the cost instead of the usual 40%. It was originally planned for occupany by English High School, but on April 23, 1973 the committee voted to use it for Girls Latin School.6 The court enjoined the committee from turning the facility over to Girls Latin.
The committee’s next case challenged the board’s plan for racial balance which had grown out of the Jaffe hearings. Noting the committee’s complete inaction since those hearings, the court rejected the challenge, putting the committee on notice that the time for testing the statute had passed, and it was time for implementation. School Committee of Boston v. Board of Education, 1973 Mass.Adv.Sh. 1315, 302 N.E.2d 916 (1973).
On November 14, 1973 a single justice of the Supreme Judicial Court ordered the committee to file with the board a detailed plan for implementation of the board’s order. On December 11, 1973 the committee filed two documents with the board, which on December 26, 1973 disapproved of portions of both and ordered that specific steps be taken to insure implementation in September of 1974. On January 4, 1974 the board filed a petition seeking judicial enforcement of its orders. On January 9, 1974 the committee represented to the single justice that it would comply and the justice accordingly adjourned the hearing until January 16, 1974, the day after the deadline for compliance. The justice on that date determined that the committee had not complied with the board’s order and ordered compliance by January 21, 1974. Board of Education v. School Committee of Boston (January 16, 1974).
On March 16, 1974 the board again petitioned the Supreme Judicial Court for further enforcement of its orders on the basis of the committee’s responses subsequent to the January 16, 1974 order entered by a single justice of the court. The court again found that the committee had not complied with the board’s orders and that it had not complied fully with the court’s order, and stated that several of the committee’s submissions manifésted a continued attempt to delay implementation of the racial balance plan. Board of Education v. School Committee of Boston, 1974 Mass. Adv.Sh. - (Mar. 22, 1974).
*420On April 17, 1974 a single justice of the court entered a further order that the committee comply with other detailed board orders. With respect to the plan for balancing Boston’s schools in September of 1974, the order specifically required that staff assignments be completed by May 1, 1974, that the staff be officially notified of those assignments by May 15, 1974, and that the safety and transportation plan for students be completed, adopted by the committee, and submitted to the board by May 15, 1974.7
At several times during the course of the proceedings in this court, the city defendants have advanced the theory that the committee has been found to be in literal compliance with the Racial Imbalance Act by the state courts; that there are restrictions in the Act and in interpretive decisions limiting the measures which school committees may take in alleviating imbalance; and that consequently, since the committee has done all that it could under the Act, the state law itself and the state defendants are responsible for racial segregation in Boston.
This theory of the city defendants must be rejected on several grounds: (1) Boston is only now on the threshold of real compliance with the 1965 statute as a result of numerous board orders and a series of unfavorable court decisions. (2) The record in this case simply will not support a finding that the city defendants saw themselves caught on the horns of the legal dilemma of choosing between fidelity to state law or to federal constitutional principles, and in good faith chose to follow state law. (3) Even if they had done so, assuming- a conflict, their choice would not have been constitutional. (4) The premise underlying the city defendants’ theory, viz., that the Racial Imbalance Act and the Constitution of the United States are in conflict, is plainly specious. As to the facial validity of the Act, we subscribe both to the Supreme Judicial Court’s conclusion of constitutionality and its observation,
“It would be the height of irony if the racial imbalance act, enacted as it was with the laudable purpose of achieving equal educational opportunities, should, by prescribing school pupil allocations based on race, founder on unsuspected shoals in the Fourteenth Amendment.” School Committee of Boston v. Board of Education, 1967 supra, 352 Mass, at 698, 227 N.E.2d at 733.
As to any constitutional attack on the Act as applied, Boston’s present racial segregation is hardly the result of the city defendants’ having complied with the state statute. Indeed it is just the opposite.
Federal Administrative Proceedings
The city defendants have also encountered administrative sanctions by the federal government for alleged violations of the Civil Rights Act of 1964: alleged discrimination on the basis of race. At issue in federal administrative proceedings are many millions of dollars in federal funds in a wide variety of educational aid programs.8
Elementary and Secondary Education Act (ESEA), Title I, 20 U.S.C. §§ 241a-244
1971— $6,105,501
1972— $8,817,298
1973— $5,928,429
Federal Assistance for Education of Handicapped Children under P.L. 89-313, Amendment to Title I, ESEA
1972— $24,000
1973— $21,000
*421School Library Resources, ESEA, 20 U.S.C. §§ 821-827
1971— $124,997
1972— $145,371
1973— $145,371
Supplementary Educational Centers and Services ESEA Title III, 20 U.S.C. §§ 841-848
1971 — $576,984 1973 — $559,494
Education of Handicapped Children, Title VI, ESEA, 20 U.S.C. §§ 871-880
1971— $41,231
1972— $33,645
1973— $89,345
Strengthening Instruction in Critical Subjects Title III NDEA, 20 U.S.C. §§ 441-444
1971— $105,029
1972— $100,914
Vocational Education, Vocation Education Act of 1963, 20 U.S.C. §§ 1241-1355, 1391
1971— $228,080
1972— $176,046
1973— $455,836
School Assistance in Federally Affected Areas, 20 U.S.C. §§ 236, 240
1971— $611,072
1972— $643,976
1973— $600,000
Education Professions Development Act, 20 U.S.C. § 1091, P.L. 90-35, B2
1972— $100,000
1973— $120,000
Grants to Strengthen State Departments of Education, Title V, ESEA, 20 U.S.C. § 862
Children of Migratory Workers, 20 U.S.C. §§ 241c (a) (b), 241e(c)
Guidance, Counseling and Testing, Title V-A, NDEA, 20 U.S.C. §§ 481-485
School Construction in Federally-Affected Areas, P.L. 81-815, 20 U.S.C. §§ 641-645
Teacher Corps, 20 U.S.C. §§ 1101-1107a
Drop-Out Prevention, 20 U.S.C. § 887
Economic Opportunity Act of 1964, Title II, Community Action Programs — Special Programs and Assistance: Head Start and Follow-Through, 42 U.S.C. § 2809
Area Redevelopment Act, Section 241, 42 U.S.C. § 2610a
Emergency School Assistance Program, 45 C.F.R. 181
Strengthening Instruction in the Humanities, 20 U.S.C. §§ 951, 952, 961
By letters dated November 30, 1971 and June 2, 1972 the Office of Civil Rights of the Department of Health, Education and Welfare (HEW), pursuant to 42 U.S.C. § 2000d-l, attempted unsuccessfully to secure voluntary compliance with the federal Act. On June 2, 1972 consolidated compliance proceedings were begun with HEW, the National Science Foundation, and the Department of Housing and Urban Development (HUD) participating. Since then federal funds in the above-listed amounts, and more, have been withheld from the public schools of Boston pursuant to 42 U.S.C. § 2000d-l. After notice and extensive pre-hearing proceedings, the hearing began on September 19, 1972 and continued until October 5, 1972. Following the filing of briefs and oral arguments, Administrative Law Judge Ring made his initial decision finding the city defendants in violation of the federal statute. In the Matter of Boston Public Schools, March 2, 1973. Judge Ring’s decision was affirmed, with minor exceptions, by the final reviewing authority in HEW, In the Matter of Boston Public Schools, April 19, 1974, which found, with HUD’s concurrence, that the city defendants have been guilty of de jure segregation.
The federal officials determined that Boston had, by persistent continuing segregative practices, intentionally created a dual school system. Included in these practices were the creation of the middle school system with feeder patterns into and out of those schools and the complementary attendance options of open enrollment and later controlled transfers with assorted exceptions. The city defendants’ practices with respect to the McCormack, Thompson, Michelangelo, Lee and other schools were found to have purposefully created segregated schools.9
The Boston School District in General
The Boston school system, whose boundaries are coterminous with those of the city, is headed by a five-member school committee elected at large by the voters of Boston. There is also a board of superintendents consisting of the superintendent, who is the chief school administrator, deputy superintendent, associate superintendents and the business *422manager. Each member of the board, which advises the superintendent, is in charge of a department. These departments include school operations, educational planning in which there is an educational planning center, personnel, staff training and development, special services, curriculum and inspection, and the business department. Each assistant superintendent is in charge of a portion of Boston known as a district. A district contains all levels of schools.
The school system is funded through a complicated statutory formula, applicable only to Boston, which allows the school committee to appropriate the great bulk of the system’s funding but requires the Boston city council to appropriate another portion upon the request of the school committee and the recommendation of the Mayor. The committee also lacks autonomy in selecting sites and contracting for the construction of new schools, as to which final responsibility lies with the Public Facilities Commission, an" independent agency of the city.
The Commonwealth of Massachusetts traditionally has exercised little authority over the day-to-day operation of local school districts. The role of the state department of education, which is under the supervision of the state board pf education, with respect to local school systems is largely supportive — providing a communication and information center, specialized services and planning. The chief administrator for elementary and secondary education within the department is the commissioner of education. The primary authority of the department is in the promulgation of educational standards: minimum school day lengths, minimum number of days in a school year, professional standards, maximum pupil-teacher ratios, ages for school attendance, nutritional standards, minimum educational standards, minimum building standards, etc. The department has two other important responsibilities, the distribution of federal and state aid, see Mass.G.L. c. 15, § 1G, and the supervision of the Racial Imbalance Act, id. at § II. One other provision should be specifically noted.
“The board shall see to it that all school committees comply with all laws relating to the operation of the public schools and in the event of noncompliance the commissioner of education shall refer all such cases to the attorney general of the commonwealth for appropriate action to obtain compliance.” Id. at § 1G.
While this division of authority between the state department of education and local school committees may be traditional, preferred, and most politically feasible, it is not required. The state has the authority, if the legislature should so enact, to take education out of the hands of local officials completely. Thus local officials derive all of their authority from the state ultimately.
Boston, one of the oldest cities in the country, retains much of its colonial layout, particularly in the downtown area. Bays and rivers and the street layout of the city have to some degree isolated some sections. The Charlestown and East Boston areas, surrounded by the Atlantic ocean and the Charles and Mystic rivers, are north of the center and have limited access to Boston proper. The South Boston area has less of a restriction on access, but remains a distinct section of the city. As in other places, political boundaries have also isolated one segment of the city. Thus, the Brighton-Allston area to the west, while easily accessible, is cut off politically from the rest of the city by the Town of Brookline except for a narrow connecting strip of Boston along the south bank of the Charles River.
The Boston school system was at one time a system of districts, subdivisions of the city, which funneled students from elementary schools through district junior high schools to district high schools. Boston has also had citywide schools for many years. These original*423ly fell into two categories, special examination schools and citywide trade schools. The special examination schools were Boston Latin and Girls Latin for college preparatory work, and Boston Technical for technical and trade work. The citywide trade schools were Boston Trade and Girls Trade, with no examination required for admission. In addition to these trade schools special programs such as auto mechanics, upholstery and electricity were offered at the various district high schools and open to students throughout the city. From this basic organization the Boston school system has evolved into a complex maze of patterns, policies and types of schools— with various exceptions to all of these— which has completely changed the original district design.
At present Boston has mismatched modes of grade progression through the school system, as follows: 8-4; 5-4-4; and 6-3-3. This system is further complicated by the possibility of transferring, or being transferred, from one progression into another. With the exception of special examination schools and trade programs, the moving of students who graduate from one school into the next level school is accomplished by feeder patterns. A feeder pattern channels students from one or more schools in one level into one or more schools at another level. Feeder patterns, as distinguished from the intra-district progression method, began to be the primary method of funneling students in 1967-68 when the grade system was changed by the introduction of middle schools. After the opening of four middle schools, a plan for the conversion of the entire system to this grade design has apparently been abandoned or post-, poned indefinitely.
Boston’s schools at the middle, junior and senior high levels now obtain students from large parts of the city. Some students, such as those in South Boston, East Boston, and Charlestown, still move through schools in those areas as if they were district schools; however, their schools also serve large numbers of students from other parts of the city.
At present three of Boston’s high schools admit students by examination; four others admit students from all parts of the city; three admit students from specified districts as well as from all parts of the city; eight admit from specified districts, except that eligibility for their vocational training programs is citywide. The junior high and middle schools have equally complex methods of filling their classrooms.
As for elementary schools, the city has a complex maze of single school and multi-school districts which vary widely in size. The elementary schools generally cannot be fairly characterized as neighborhood schools. District lines require some students to travel many blocks to a school when there is another school much closer, but on the other side of that district line. In multi-school districts, students at least when enrolling for the first time have a complete option as to which school to attend and often walk past one en route to another. Some elementary schools, community schools and magnet schools are further affected by other programs. Community schools have been designed and constructed with a view to involving the nearby community and their facilities are available to adults after school hours. Some have swimming pools and other athletic facilities and all provide meeting places for community organizations. There are nine such elementary schools and one junior high, the Cleveland. Magnet schools, of which there are four, have districts which are smaller than would be normal for their capacities. The idea is that students from outside of the district and outside of the city will be attracted to attend them in order to take special courses. Two, Trotter and Hennigan, were built in non-white neighborhoods. Whites attracted to these schools were supposed to fill vacant seats after students residing in the district were seated.
*424Three schools, an elementary (Trotter), a junior high (Lewis) and a senior high (Copley Square), comprise a so-called model demonstration subsystem. Until federal funds were cut off, they received approximately 30%, 40% and 50% respectively of their financing from the federal government under Title I of the Elementary and Secondary Education Act, 20 U.S.C. §§ 241a-244, because of special programs designed to aid disadvantaged children of parents receiving Aid to Families with Dependent Children. Eligible white students from outside Boston, from between 15 to 20 suburban communities, are bused at no charge to and from these schools.
Approximately one-third of Boston’s students, a large majority of whom are in high school, use buses or other public transportation to travel to and from school. Approximately 3,000 elementary students are transported at city expense, most of whom attend schools over a mile away from their homes. In Charlestown some elementary students who live less than a mile from school are bused for safety reasons. Other elementary students are bused several miles, e. g., from the Dearborn district in Roxbury to the North End and East Boston; others from the South End to Brighton. The three examination high schools, sometimes called the “elite schools”, were served in the school year 1971-72 by a combined total of 63 buses on 35 routes. Many other students travel between distant parts of the city.
Two programs for transporting black pupils to predominantly white schools were started by black parents as private projects and later financed by the state. One, called Exodus, buses blacks within the city; at its peak in 1969 it handled 1,100 pupils, a figure reduced to 170 in 1972. The other, which started in 1966 and is called Metco after its sponsor the Metropolitan Council for Educational Opportunities, takes black pupils to schools in the suburbs; the number transported has increased steadily and now totals approximately 1,650.
The Degree of School Segregation
The Boston public schools are characterized by heavy concentrations of black pupils in some schools and heavy concentrations of white pupils in other schools. As of the 1971-72 school year, during which this action was filed, approximately 96,000 students were enrolled in the system, of whom about 59,300, or 61%, were white; 30,600, or 32% were black; and 6,500, or 7%, were other minorities. This overall ratio, which still exists, is far out of line with the ratios in most of the system’s schools. Eighty-four percent of Boston’s white students attend schools that are more than 80% white; 62% of the black students attend schools that are more than 70% black. At least 80% of Boston’s schools are segregated in the sense that their racial compositions are sharply out of line with the racial composition of the Boston public school system as a whole. Johnson v. San Francisco Unified School District, N.D.Cal.1971, 339 F.Supp. 1315, 1329. See also Swann v. Board of Education, 1971, 402 U.S. 1, 25-26, 91 S.Ct. 1267, 28 L.Ed.2d 554. Racial segregation permeates schools in all areas of the city, all grade levels, and all types of schools.
Boston has eighteen high schools. Only one, Boston High, which features a work-study program, reflects the racial composition of the student population in the Boston school system. Only one other, Brighton High, is within 10% of that makeup. Five high schools are in excess of 90% white. Three others are 85% white. Two high schools are 90% black and have a white student population of 2% or less. Four others have a more than 50% black composition: 52%, 66%, 66% and 75%. The two remaining high schools have a combined black and other-minority enrollment in excess of 55%. This pattern persists at specialized schools: Boston Latin and Girls Latin, special examination schools for college preparatory work, are 93% white and 89% white respectively. The trade schools in Boston are similarly segregated: Boston Technical, 84% *425white; Boston Trade, 66% black; and Girls Trade, 75% black.
A similar pattern obtains in the junior high and middle schools. Boston’s four middle schools are 94% black, 86% black, 72% black, and 63% black. Of fifteen junior highs, six are over 90% white, one is 88% white, one 81% white. Three junior highs are 95% black. Only the Edison and Taft junior highs approach the racial makeup of the school system, but they miss that by 10% and 20% black shortages respectively.
Boston has ten elementary schools ending in grade eight. Five have enrollments greater than 82% white; one is 94% black; another, a combination of 78% black and 15% other-minority. Two others are more than 60% black and other minorities. The other 140 or so elementary schools in the city, in which the highest grades are fifth or sixth, have similar compositions. Sixty-two are less than 5% black; thirty-two are 85% or more black. The remaining elementary schools also have attendance ratios sharply out of line with the racial makeup of the school system. Only five of these approximately 140 elementary schools, Marshall, Taylor, Barrett, Stone and Curley, have a racial composition within ten percent of the citywide 61:32 ratio of whites to blacks.
Teachers are also segregated. Seventy-five percent of Boston’s black teachers are in schools more than 50% black. Eighty-one schools have never had a black teacher.
The Boston public school system is thus characterized by racial segregation. The defendants do not dispute this central fact. The dispute, rather, is how the schools have become and remained that way. The court’s primary task is to determine whether the defendants have intentionally and purposefully caused or maintained racial segregation in meaningful or significant segments of the Boston public school system, in violation of the Fourteenth Amendment. Keyes v. School District No. 1, Denver, Colorado, 1973, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548; Swann v. Board of Education, 1971, 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554; Brown v. Board of Education, 1954, 347 U.S. 483, 74 S. Ct. 686, 98 L.Ed. 873; Davis v. School Dist. of Pontiac, Inc., E.D.Mich.1970, 309 F.Supp. 734, aff’d 6 Cir. 443 F.2d 573, cert. denied, 1971, 404 U.S. 913, 92 S.Ct. 233, 30 L.Ed.2d 186. In making this determination, the court has analyzed the defendants’ conduct in six principal areas, to be discussed separately, as follows, (1) facilities utilization and new structures, (2) districting and redistricting, (3) feeder patterns, (4) open enrollment and controlled transfers (5) faculty and staff and (6) vocational and examination schools.
I
Facilities Utilization and New Structures
The plaintiff and the defendants presented much evidence pertaining to overcrowding, the use of portable facilities,, and new annexes and schools. The crucial evidence consisted not only of the decisions on the use of school facilities but also of the complex contexts in which these decisions were made. Thus the planning, timing, educational justifications and other factors all pertain to intent or purpose. The effect of school officials’ actions is more readily apparent. The first relevant facet of the situation is the crowded condition of some schools.
Overcrowding
The Boston system is oddly riddled with groupings of school facilities at each grade level that are extremely overcrowded. These overcrowded schools stand in stark contrast to other facilities at all grade levels which operate far under their capacity. Generally, the overcrowded schools are predominantly white and the under-utilized schools are pre*426dominantly black. The following tables for the school year 1971-72 demonstrate this condition with several of the more extreme situations in the high schools, junior highs and one middle school:
Racial Makeup
Overcrowded Schools * B% 0M% W%
Charlestown High 2.0 6.6 91.4
Dorchester High 52.2 1.1 46.8
Hyde Park High 15.3 .7 84.1
Roslindale High 5.1 1.6 93.3
South Boston High 0 .7 99.3
Cleveland Jr. High 7.1 1.8 91.2
Gavin Jr. High 1.8 1.8 96.4
Rogers Jr. High 3.3 .4 96.3
Underutilized Schools *
Boys Trade 66.5 6.9 26.6
English High 66.7 8.5 24.8
Girls Trade 74.9 5.1 20.0
Girls High 91.7 5.8 2.5
Jeremiah Burke High 89.0 10.1 .9
King Middle School 94.4 5.4 .2
Timilty Jr. High 95.0 3.1 1.8
In some instances this overcrowding has been extreme, e.g., South Boston High, all-white, was overenrolled by 676 students in the 1971-72 school year. In contrast, Girls High, 92% black was underenrolled by 532 places.
These patterns are also observable at the elementary level.
Available Seats
Identifiably Black 1971-72 1970-71 1969-70
P. Brooks 108 88 107
Q. Dickerman 75 66 45
Mackey 83 51
Bancroft 60 66
Williams 73 70
Carter Temporary 79
Dudley 156 131
Bacon 79 60
Dillaway 145 136
Hale 55 58
J. Bates 52 51 68
Lincoln 153 47 143
Enrollment Over Capacity
Identifiably White 1971-i 1970-71 1969-70
Everett 126 138 153
Russell 27 48 18
Clap 68 76 52
O'Hearn 54 48 57
Hemenway 128 111 98
Mather no 123 134
Southworth 16 42 74
Cushing 29 95 42
Chittick 108 Longfellow 102 122 121 112 116
Overcrowding is educationally damaging partly because it strains the capacity of core facilities such as libraries, classrooms, science laboratories and trade shops. The adverse educational effects of overcrowding within a school facility are so obvious that we shall not discuss it further except to note that by words and deeds the defendants10 often knowledged that fact. However, the defendants have responded on at least one occasion to the problem of overcrowding with actions contrary to sound educational practice. In alleviating overcrowding at Cleveland junior high, 91% white, students were assigned to the already overcrowded and relatively distant white South Boston High. There were closer schools with available seats but these schools were identifiably black. Similarly, when it would have reduced racial segregation, the defendants ignored other opportunities to decrease overcrowding by altering school assignments, viz., the group assignment and busing of black students to the Weld school, nearly every use of portable classrooms and the opening of the Lee school.
While it would not always have been necessary, busing was a viable alternative to overcrowding. Boston buses several thousand school pupils. In June 1971 defendants proposed busing 4,000 inner city black students as much as fifteen to twenty miles to suburban schools in exchange for an equal number of white students from the suburbs. The defendants have not felt constrained by the busing restrictions in the Racial Im*427balance Act, Mass.G.L. c. 71, § 37D, which they correctly viewed as constraining state officials only.
Some evidence tended to show an explicit racial reason for not relieving overcrowding. For example, Deputy Superintendent Meagher did not consider assigning students from overcrowded white schools to black schools with available space because he “thought it would create a problem” of white parents protesting. Assistant Superintendent Griffith did not favor such assignments because he knew that white parents were opposed to such transfers.
On several occasions defendants pursued a policy of reducing overcrowding when it would not affect the racial composition of the schools involved, e.g., changing the boundary lines of the Cleveland and Campbell (now King) districts in 1959 and the Boardman district in 1964, housing overflow students from English High in the Roosevelt and Edison Schools from 1964 until 1970, assigning blacks from overcrowded schools to the Weld from 1970 until 1972, and transporting the entire Kennedy school fourth grade to the Boardman in 1969 due to overcrowding at the Kennedy.
Finally, the defendants deliberately dragged their feet in formulating plans to lessen overcrowding as well as racial imbalance generally. They constantly delayed the presentation of plans requested by the state board until the last possible moment. For example, a plan to balance high schools in South Boston, East Boston and Charlestown was under discussion at a school committee meeting on June 1, 1971 when the superintendent stated, referring to the commissioner of education, “If you want my quick reaction to it, I’d hold this plan in my back pocket until he demands it.”
Thus the defendants have recognized the educational hazards of overcrowding but have acted inconsistently on the basis of the race of the students being harmed. The following subdivisions deal specifically with affirmative acts of the defendants related to overcrowding which we find intentionally created or maintained racial segregation.
Portable Classrooms
Portable classrooms are semi-mobile, one classroom buildings which are designed for quick temporary service without large capital expenditures. Boston has used them since the 1960’s to alie-, víate the overcrowding of some schools. The use of these facilities and the defendants’ contradictory statements regarding their use are highly probative.
In 1967 the city placed 50 of these units at various schools in the city. The following tables demonstrate how they were used to alleviate overcrowding at predominantly white schools:
Number of Schools Percentage Black Number of Portables
* 10 0- 5% 24
4 5- 15% 20
0 15- 30% 0
2 30- 50% 5
1 50- 70% 1
0 70-100% 0
In 1972-73 there were 46 in use and their location was as follows:
Number of Schools Percentage Black Number of Portables
6 0- 5% 26
4 5- 15% 8
1 15- 30% 3
2 30- 50% 5
0 50- 70% 0
2 70-100% 4
Portables were used to alleviate overcrowding when available nonsegregative methods, such as changing district lines and reassigning students, could have achieved the same result. The use of these portables perpetuated the racial concentrations present in both the overcrowded schools and the schools to which displaced students might have been sent.
The city defendants’ policy regarding the use of portables has shifted back and forth, depending on whether their use was proposed as a means of reducing segregation or of correcting overcrowding at predominantly white schools. The Kiernan Report recommended their use for the former pur*428pose, but in June 1965 the defendants decided against it because portables had been proved to be, in Superintendent Ohrenberger’s phrase, “educationally undesirable.” Simultaneously, however, their use for the latter purpose was under consideration and on September 13, 1965 the committee voted to place portables in South Boston. According to the then chairman, a reason for this was that “there would be no question regarding the racial imbalance question.” In February and June 1966 the state task force on racial imbalance criticized plans submitted by the committee for failing to propose, among other things, the use of portables to reduce racial imbalance. On June 2, 1966 the committee again rejected the idea, with various members raising objections of expense and parental opposition. But on March 6, 1967, the committee approved the use of 38 new portables to control overcrowding. Again in 1968, 1969 and 1971 the school committee declined to follow state board proposals for the use of portables in the city’s racial imbalance plans for those years.
New Facilities
Boston has increased its seating capacity by the construction of new schools, the building of annexes and the acquisition and conversion to schools of such diverse structures as a church, a synagogue, an automobile showroom, a bowling alley and a bathhouse. The overwhelming effect of most of these projects has been to increase racial segregation. The following table shows the racial composition of these new schools or annexes upon opening:
Racial Makeup upon Opening
Facility School Year Opened * B% 0M% W%
Lewis Senior Annex 1967 ('67-'68 only) 98.6 0 1.4
Garrison Annex 1970 98.4 1.6 0
Paine Annex 1970 91.5 1.6 6.9
Weld 1970 (closed '72) 87.1 5.8 7.1
Bradford Annex 1969 81.2 9.4 9.4
Lewis Middle Annex 1970 ('70 — '71 only) 79.1 .9 20.0
Tobin Annex 1969 74.6 14.5 10.9
Carter Temporary 1971 72.6 4.7 22.7
Kennedy Annex 1969 ('69 — '70 only) 60.7 15.2 24.1
Copley Square High 1968 56.3 3.0 40.7
Dorchester Annex 1970 53.1 .5 46.4
Haley 1971 47.2 8.7 44.1
Boston High 1968 39.8 2.0 58.2
Hamilton Annex 1970 11.3 1.0 87.6
Georgetown Kd. 1970 (closed '72) 3.0 .4 96.6
Charlestown Annex 1971 2.0 6.6 91.4
Sumner Annex 1969 1.9 .9 97.2
Dean Í972 1.0 3.0 96.0
L Street Annex 1970 .7 2.3 97.0
Hart 1972 .5 8.3 91.2
Contrary to the command of the Racial Imbalance Act that “[t]he prevention or elimination of racial imbalance shall be an objective in all decisions involving the selection of new school sites”, Mass.G.L. c. 71, § 37C, and that plans should detail “proposed additions to existing school buildings”, § 37D, the defendants took the position that the effect on racial imbalance was irrelevant to the planning of several of these new facilities. Significantly, many of the annexes that were not subjected to state scrutiny, because constructed without state financial aid of any sort, opened identifiably black. If state aid had been applied for when these facilities were being planned, the state board would routinely have in*429quired about their probable impact on racial imbalance. Evidently the defendants preferred not to have to respond to that type of inquiry. From the percentages and circumstances already described, and the specific episodes which follow, it is apparent that, in the matter of facilities utilization and new structures, the defendants were covertly resisting the elimination of racial imbalance and endeavoring to perpetuate racially segregated schools.
(a) The Weld School. This school, located in Roslindale, where 98% of the residents were white, was opened in September 1970 after purchase from the Archdiocese of Boston for which it. had been a parochial school. For two school years, 1970-71 and 1971-72, it was used to relieve an overcrowded situation in the 90% black Bradford-Walcott district a mile and one-half away. There approximately 130 non-white students and 11 white students were selected mostly by lot and assigned to 'the Weld, where they were joined by 14 students from the vicinity who suffered from visual handicaps. Weld opened 87% black, 6% other-minority and 7% white. After commencement of the instant action and objections by the state board, HEW and representatives of the black community, Weld was closed. It may be, as contended by defendants, that it was planned from the start as a temporary facility. This does not explain, however, why none of the regular, i.e., nonhandicapped, students from the neighborhood of Weld were assigned to attend it. Moreover, nearly all of the 57 first and second grade black pupils might have been assigned, with no greater inconvenience, to vacant seats in predominantly white schools in the Sumner and Longfellow districts. Parents of the black students were given no choice about their children being bused distances of from one and a half to two miles. Theretofore the defendants had adhered to a policy of busing only on a voluntary basis. The first attempt by the defendants to ascertain the attitude of black parents toward busing to Weld came in April 1972, three months before the school was closed.
The defendants have argued that the reason for transporting the BradfordWalcott students to the Weld school was to keep existing friendships intact. This argument must be rejected because there is no evidence that this reason was considered by the defendants before busing was undertaken; the evidence supporting it appears exclusively in documents prepared after HEW and the state board charged the defendants with discrimination in relation to the Weld school.
(b) The Hennigan School. This new elementary school was proposed in 1967 as part of Boston’s first stage racial imbalance plan and was built with 65% state financing as a replacement for the Jefferson school which had a capacity of 500 students. One wing of the school was opened in 1971 with 302 students, 65% black, 10% other-minohky and 25% white. The school had a capacity of 1,080 students and was planned to accommodate students residing in the nearby Heath Street housing project, 99% black, and white students to be recruited from outside the district. The defendants made no efforts to recruit white students for the 1971 opening.11 As a consequence of its location and defendants’ action, the school was identified as tailored for black students. Thus in the following year, the defendants distributed 50,000 recruiting brochures but garnered only 50 white recruits. A three-day “open house” in June 1972 was the last effort made by the committee. The entire school opened in the fall of 1972, 62% black, 13% other-minority and 25% white. There is no justification for the defendants’ reli.ance upon obtaining white volunteer students rather than assigning white stu*430dents to the school. They argue now that failure of the efforts to attract white volunteers to Hennigan could not have been foreseen because, after all, white students voluntarily enrolled in Trotter.12 But that was in 1969. Times had changed and local opposition to racial balancing was on the rise. In 1971 negligible numbers of white students in the Fifield and O’Hearn districts had volunteered to attend Lee. The analogy also limps because of essential differences between Hennigan and Trotter, which was the elementary school component of the model demonstration subsystem to which white children of AFDC recipients were bused from all over the greater metropolitan area. The attendance of black students at the Hennigan school was mandatory; -its racial segregation upon opening could not have come as a surprise.
(c) The Lee School. This school was also proposed as part of the first stage racial imbalance plan and was planned to replace kindergarten through grade 5 of the Champlain, Whittier and Nightingale schools. Because its construction was supposed to contribute to racial balance, the state paid 65% of its cost of over $7,000,000 pursuant to Mass.G.L. c. 15, § II. From the beginning the state board pressed the school committee for assurances that the Lee would open racially balanced. On June 1, 1971 the school committee decided to include within the new Lee district approximately 350 white pupils then attending the Fifield and O’Hearn schools but gave them the option of attending their former schools. This was done in the face of warnings by Mr. Coakley of the city’s Educational Planning Center (EPC) that the option would produce an imbalanced school. On July 12, 1971 the committee was presented with hard evidence that the Fifield-O’Hearn option would produce an imbalanced Lee; it was told that no O’Hearn parents at all and the parents of only 37 Fifield pupils had indicated that they would send their children to the Lee. To compel reconsideration, the state board threatened to withhold state funds. So on August 23, 1971, barely two weeks before the beginning of school, the committee eliminated the Fifield-O’Hearn option by a 3-2 vote. Significantly neither the principal of Lee school, Miss Kelley, who had worked many months to assure a successful opening, nor the Fifield and O’Hearn parents were told of the August 23 vote.
Opening day at Lee school, September 8, was chaotic. Approximately 200 black students from the predominantly black Franklin Field housing project nearby illegally registered by giving false addresses and occupied seats. The Fifield and O’Hearn pupils registered at their former schools, refusing to go to Lee. Confronted with this situation, the committee on September 17 scheduled a public meeting at the O’Hearn school on September 21, and, with one member switching his vote, reinstated the Fifield-O’Hearn option and allowed the black sit-in pupils to remain at Lee. The finding of the Superior Court, to which the defendants are bound, was that “this vote was a measure of expedience in the face of public furor.” Boston School Committee v. Board of Education, Suffolk Sup.Ct., No. 94254, at 42. We add that, to a large extent, the crisis was of the committee’s own making.13
The defendants acted in such a way as to lead all parties concerned to believe that it would not compel the Fifield and O’Hearn pupils to attend Lee. It did virtually nothing to encourage the parents of those students to send them to *431Lee, even when it became inevitable that Lee would open racially imbalanced if those students were not in attendance. The intent of the defendants was apparent all along: they never intended to take the initiative in seeing that Lee would open racially balanced. The segregated consequences were clearly foreseeable.
Parenthetically, assignment of Fifield and O’Hearn pupils to Lee would have alleviated overcrowding at those schools. Educational considerations, thus did not justify the defendants’ failure to advance integration at Lee. Moreover, the routes from Fifield and O’Hearn were safe and no transportation was necessary. Safety considerations thus did not explain the school committee’s actions.
The defendants have sought to justify their capitulation of September 21 on the ground that the black pupils who illegally registered at Lee were adamant about staying and the Fifield-O’Hearn option was reinstated to prevent overcrowding at Lee. This proffered theory is pure rationalization and completely unsupported by contemporaneous evidence of the reasons for the committee’s action. Furthermore, permitting the Fifield and O’Hearn pupils to return to their former schools merely perpetuated overcrowding at those schools. Finally, the defendants’ suggestion that, but for the illegal registration of 200 black students, they might have recruited white students from other sections of the city to attend Lee overlooks the fact that the recruiting season had ended as of opening day.
(d) New English High School. In September 1973, English High, which is located in the Back Bay Fens section of the city near several colleges, hospitals and museums, was moved next door into a new $24,000,000 facility ten stories high, the first new high school opened in Boston in 35 years. The transfer did not materially affect racial segregation. However, it came only after a single justice of the Supreme Judicial Court had nullified a 3-2 vote of the school committee awarding the facility to another high school, Girls Latin, and the events leading up to it bear on the issue of defendants’14 intent.
The new building would accommodate roughly 2,000 students. The enrollment at both English and Girls Latin were almost identical, just short of 1,200 students each. The new building was designed and constructed to house English, with special facilities for courses in business, merchandising, electronics, carpentry, motor repair, and other courses which would not fit the Girls Latin college preparatory curriculum. The new building was constructed with 65% state financial assistance under plans formulated in 1967 and 1968 to reduce racial imbalance. The single justice found on all the evidence presented to him that the defendants made a commitment to the state board that English would be housed in the new tower facility upon its opening, but intended by their vote of April 23, 197315 to exclude English from the tower facility upon its opening. Substantially the same evidence was presented to this court when, after trial the case was reopened on plaintiffs’ motion, and we agree with the state court’s findings.
Why did the school committee try to renege on its commitment to the state? Plaintiffs contend that it was because the composition of the student body at English had changed from approximately 20% black in 1967, when *432the new building was planned, to over 80% black in 1973 when, on the eve of its opening, it was awarded to Girls Latin, 89% white. Plaintiffs have requested a finding that the committee placed the black students at English “in the position of what may be described as second-class pupils”, citing Lee v. Macon County Board of Education, 5 Cir. 1971, 448 F.2d 746, 754 n. 12 quoting Brice v. Landis, N.D.Cal.1969, 314 F.Supp. 974, 978. Plaintiffs’ requested finding raises a question whether blacks were denied educational opportunity, cf. Hobson v. Hansen, D.D.C.1967, 269 F.Supp. 401, aff’d sub nom. Smuck v. Hobson, 1969, 132 U.S.App.D.C. 372, 408 F.2d 175, which we do not reach because the committee was enjoined by the state court from carrying out its vote. The issue before us is primarily one of segregative intent. On that issue, there is evidence that the defendants intended to keep English’s student body heavily black, and we so find.
As explained in the division of this opinion entitled “Feeder Patterns,” the rapid change from 20% to 80% black in the racial composition of the student body at English, which is undistricted, i. e., with students from all parts of the city, was attributable in large measure to feeder patterns knowingly implemented by the defendants. The new facility had capacity for approximately 700 students in addition to the number attending the old English; the committee had advised the state board that this excess capacity would be used for white students, thus balancing it racially; and the headmaster and staff at English had developed courses and projects which would probably have made this goal attainable, especially if combined with feeder pattern changes which the headmaster recommended repeatedly but to no avail. After the committee’s vote awarding the new facility to Girls Latin, 89% white, the ancient building which had been occupied by English was razed and the defendants made no plans for housing English other than in the building in Codman Square which had been occupied by Girls Latin. True, as argued here by the defendants, the students at English were never formally assigned to the'building which would have been vacated by Girls Latin; but as a practical matter there was no alternative —the only word given the English headmaster by his superiors was that English would be moved to that building. The building at Codman Square, which incidentally is almost as antiquated as the old English building, is located on the border of a heavily black neighborhood. Had English been sent there, it would have been more firmly identifiable as tailored for the education of blacks.
II
Districting and Redistricting
Districting denotes the drawing of geographical boundaries which will determine attendance zones for a given school facility. In Boston, districting is a misnomer as applied to high schools because their attendance is either citywide or determined by feeder patterns geared to schools rather than to geographical areas. Even high schools located in definite sections of the city, e. g., Brighton and Roslindale, receive students on the basis of the intermediate schools from which they have graduated. Feeder patterns thus control the composition of the student bodies at secondary schools; and, while they serve the same purpose as the drawing of geographical boundaries, they will be discussed in subdivision III. Intermediate schools are all districted geographically, although five of them also receive students by way of feeder patterns and there is at least one instance of overlapping intermediate school districts. The student bodies at primary schools are determined entirely on the basis of residence. Each level of schools, intermediate and primary, obviously requires districting of it own.
Districting has much less to do with what school a pupil will attend in Boston than it may in other communities. This is because of practices which cut across *433district lines, such as the controlled transfer policy with its several loopholes, feeder patterns and specialized schools. Another reason is the existence of multi-school districts at the elementary level which are administered by a district principal and in which there are as many as six separate buildings in different parts of the district. Whether these various features are unique to Boston was not indicated. Surely, however, they are not characteristic of most of the public school systems in Massachusetts.
The defendants have not made districting changes which brought about increased segregation where there was none before, with the exception of districting changes pertaining to certain fifth and sixth grades giving advanced work classes leading to the three elite schools, which are too involved to warrant explanation here.16 However the defendants have made districting changes for the purpose of perpetuating racial segregation, and findings will be made about them. The bulk of the findings in this subdivision will describe extreme disparities in the racial composition of adjacent districts and the defendants’ uncompromising resistance to redistricting when it would have reduced racial segregation. Year after year the defendants rejected proposals for redistricting carefully drawn with a view to lessening racial imbalance while at all times displaying awareness of the potential racial impact of their actions. Despite intense pressure from private groups and the state board, including the withholding of millions of dollars in state financial assistance, the defendants defeated or evaded successive redistricting proposals, for the purpose of perpetuating the existing segregated system.
In the Dorchester section of Boston, district lines at the intermedicate school level preserve a high degree of racial segregation in adjacent districts. The following findings pertain to a cluster of eight districts, six for junior high schools, one for an elementary school ending in grade eight, Parkman, and one for a middle school, Thompson, whose district overlaps portions of the Lewenberg district, 95% black, and the Wilson, 81% white. The Holmes district, 95% black, whose eastern and western district lines are near the edges of the “black boomerang”, is adjacent on the east to the Cleveland, 91% white.' To the south of the Holmes and Cleveland districts are the Lewenberg district, 95% black, the Wilson, 82% white, and the Thompson, 63% black. The district for Thompson, a middle school, is completely surrounded by the Lewenberg District 97% black, and the Wilson 82% white. Graphically the area is depicted roughly as follows:
[See following illustration]
*435These districts shown in the map cover an area about three miles from east to west and two miles north to south. Their configuration results in nearly the maximum possible amount of racial isolation at these schools. Only small sections of the district lines coincide with natural boundaries, Blue Hill Avenue and a railroad track, and the latter runs either overhead or under street level. A slight boundary change between the Holmes and Cleveland districts would decrease racial imbalance at both schools and would also eliminate the situation now existing whereby pupils who are within a few blocks of Holmes, 95% black, must attend the relatively distant Cleveland, 91% white. Portions of the Wilson district, 81% white, are closer to Holmes junior high, 95:% black, and Lewenberg, 95% black, than to Wilson, 81% white. Portions of the Lewenberg district are as close to Wilson junior high as to the Lewenberg, and other portions of the Lewenberg district are almost as close to Rogers junior high, 96% white, or Irving, 88% white, or even to Parkman, 83% white, as they are to Lewenberg. Parts of the Irving district are three times as close to Lewenberg junior high as to Irving. The feasibility of a Lewenberg to Rogers or Irving switch is corroborated by the school committee’s ad hoc decision in 1972 which allowed several white elementary school graduates who lived in the Lewenberg district to attend the predominantly white Irving or Rogers junior highs rather than the predominantly black Lewenberg. The Thompson middle school district, 63% black, was created in 1968 with the conversion of an elementary school and was superimposed upon the predominantly white Wilson and the predominantly black Lewenberg. The effect of this districting was to provide students residing in the eastern part of the Thompson district with the option of attending Wilson Junior high, 81% white, and those in the western part of the Thompson district with the option of attending Lewenberg junior high, 95% black.17
The elementary schools in a portion of Boston stretching from just south of South Boston, through Roxbury and into Dorchester are districted with a similar effect: the predominantly black areas are cut away from the predominantly white areas. The following map shows how the district lines weave in and out with this effect; the broken line depicts the dividing line between the white and black elementary districts.
[See following illustration]
*437As shown in the map, on the west lie predominantly black schools; the easternmost edge of their districts coincides with this dividing line. Noticeably, the predominantly white schools to the east are located some distance away from the dividing line. The single exception to this pattern of separation is Champlain, 94% black, and that elementary school is practically on the district line. There are no natural barriers which explain this north-south dividing line; there is a railroad line which coincides with parts of it, but the tracks are either elevated or below street level. Moreover, the same railroad line runs through the middle of at least six elementary school districts in that vicinity: Northcross, Gibson, Woleott-Bradford, Tileston, Chittick and Lee.
Racial segregation of elementary students is facilitated by another feature of the city’s elementary school districts, viz., multi-school districts. Of such districts, only six have significant numbers of students with different racial backgrounds and they are segregated according to race at particular schools within the districts. This is shown by the following table in which the racial percentages are for the school year 1971-72:
Other White Minority Black
Parkman District
Parkman 82.6 9.2 8.1
Spaver 91.0 .9 8.1
Haley 44.1 8.7 47.2
Sumner District
Sumner 95.7 1.3 3
Philbrick 99.5 .5 0
Sumner Annex 93.5 .5 6
Irving Colony 97.2 1.7 1.1
Weld 44.7 3.7 81.7
Conley 99.6 .4 0
Mendell District
Fuller 72.4 16.5 11
Mendell 33 39.8 27.2
Roosevelt, T. 11.8 57.1 31.2
(Annex) .5 3.2 96.2
Marshall District
Marshall 59.0 5.4 35.6
Stone 53.5 4.4 42.1
Champlain 1.1 4.8 94.1
White Other Minority Black
Prince District
Prince 40 14.7 45.3
Faneuil 75.6 8.5 15.8
Perkins .8 15.2 84.0
Milmore 53.1 27.2 19.7
Rice-Franklin District
Bancroft 57.9 11.3 30.8
Mackey 15.2 47.1 37.7
As the table demonstrates, these multi-school districts contain distinct racial clusters. Of the six schools in the Sumner district, five are in excess of 90% white and the sixth is 81% black. By contrast, in the Mendell district the whites are noticeably concentrated in the Fuller school, while the other two schools are predominantly black and other minority. The schools in the other districts present slightly less extreme examples, but the pattern is evident.
Redistricting, in the sense here discussed, must be further defined. First, it refers to redrawing of old boundaries for old schools and is to be distinguished from new districting necessitated by the construction of new schools or by the conversion of one type of school to a different type, e. g., a grammar to a middle. Secondly, it is a term used only with reference to primary and intermediate schools. Secondary schools have districts only derivatively, i. e., their student bodies are generally determined by the districts of lower level schools whose graduates are routed to them in accordance with feeder patterns, which are discussed in the next division of this opinion. The redistricting of high schools is accomplished, in effect, by assigning and changing feeder patterns. Thus in the present context redistricting means changing attendance zones of existing elementary and intermediate schools.
On several occasions the school committee redistricted when racial balancing was not involved: in 1959 the boundary lines of the districts for Cleveland and Campbell (now King) junior highs were altered for administrative reasons; between 1968-64 and 1969-70 the districts for Roosevelt junior high in Roxbury *438and Edison junior high in Brighton were changed to house an overflow of students from English high; and in 1969-70 the Boardman district was expanded to accommodate the entire fourth grade at Kennedy due to overcrowding. Between 1954 and 1966, the school committee made sixteen other districting changes.
On at least one occasion the committee redistrieted with a view to perpetuating racial segregation. This occurred in 1968 when certain streets in the Cleveland junior high district, 91% white, were transferred to the Russell junior high district, 85% white, and to the South Boston high, 99% white. Apby the change, which was occasioned by overcrowding at Cleveland. At the time both South Boston high and Russell were already overcrowded and were further away from Cleveland than other schools with available seats, namely, King with 295 available seats, Burke with 20 and Girls High with 568. However, these underutilized schools were predominantly black: King, 94%; Burke, 89% ; and Girls High, 91%. Assistant Superintendent Griffith testified at the trial that redistricting of the King district to include students from the overcrowded Cleveland would “certainly” have been feasible, but he did not recommend it because he “knew the attitude of the people in the area” and “white parents did not wish their children in schools such as the King and the Holmes, and possible in the Timilty”, all of which are predominantly black. Nor would safety hazards have been involved in such a redistricting, at least any more than those already' encountered by students attending King.
On no occasion have the city defendants redistricted for the purpose of eliminating racial imbalance, although they have many times been shown the way. The Kiernan Report, filed in April 1965, proposed specific redistricting and was largely ignored by the school committee. Following the city’s first census pursuant to the Racial Imbalance Act, Mass.G.L. e. 71, § 37D, the board found that racial imbalance existed in Boston and notified the school committee in writing, thereby placing the committee under a legal duty to file a plan which “shall detail the changes in existing school attendance districts . ” The committee’s first plan failed to include any provisions for redistricting and was rejected by the board voted to hold in escrow state other reasons. On April 12, 1966 the board voted to hold in escrow state funds due Boston until a revised plan was filed and accepted. In early 1966 the board engaged the services of the Joint Center for Urban Studies of Harvard and MIT, which prepared several alternative redistricting proposals, eight for elementary schools and three for intermediate schools.
In drawing up the proposals the Joint Center conferred regularly with school committee personnel, including the deputy superintendent of schools. Each proposal would have substantially lessened segregation in the parts of the city shown in the maps, ante, and each was designed to meet the following requirements :
1. No district should extend more than about one-half mile from the school for grades 1-3, or three-fourths mile from the school for grades 4-6. For junior high schools the maximum is 1}4 miles, unless public transportation is available.
2. Each district should have a compact shape.
3. No child should walk past another school serving his own grade on his way to his assigned school.
4. No more than 30 children per academic classroom should be allocated to any elementary school involved in the redistricting, and no more than 32 per academic classroom to a junior high school. For schools having only minor additions to their districts, the official vacant seat count has been used in lieu of this requirement.
*4395. No white child should be required to change to a formally imbalanced school unless the reassignment would eliminate racial imbalance in that school, bringing it to 45 percent nonwhite or less.
Safety was considered. The proposals did not affect kindergartens and did not involve any busing except the use of public transit by some junior high students. They were discussed by the committee at a meeting on June 2, 1966 together with the general problem of steps which might be taken to obtain the release of the state funds being withheld. The transcript of the meeting runs 125 pages and was received in evidence, as were transcripts of many other committee meetings. A majority of the committee were set against redistricting of existing school districts in any form, for various expressed reasons, mainly that it would accomplish relatively little in alleviating racial imbalance, which was due mainly to segregated housing patterns and could never be eliminated anyway, and that transferred faculty18 and parents of redistrieted students would be bitterly opposed. There was considerable talk about distances between schools; dangers at street crossings, the educational unsoundness of overcrowding, etc., as if the Joint Center proposals had ignored these factors. But the basic issue was, in the words of one member, “Either you believe in redistricting or you don’t believe in it.” At the meeting a strategem evolved of trying to satisfy the state board with a plan to combat-racial imbalance which would not include redistricting of existing schools, but would confine districting decisions to newly constructed schools. The committee agreed to propose a four-point plan featuring new schools, an expansion of the Meteo program for exchanging black students from the city for white students from suburban schools, an open enrollment policy and compensatory programs including “sociological student-teacher exchanges” and “multi-ethnic textbooks.” Toward the close of the meeting, the chairman, who believed that the funds withheld by the state would never be released unless the city’s plan included some provision regarding redistricting and had kept urging “a step forward, a good faith. indication that we intend to comply with the law as they [the state board] interpret it”, made a motion that the superintendent be directed to review the Joint Center package and “extract from it those recommendations which you can live with and which are workable and will have the effect of” minimizing racial imbalance. The motion lost by a vote of 3-2.
The four-point plan, somewhat modified, was formally adopted by the committee at its next meeting and forwarded to the board on June 13, 1966. The board rejected it on June 28. The committee resubmitted it, unchanged, on July 6. The board rejected it again on July 26. The lines had been drawn and, on the matter of redistricting, the committee and the board have been at swords’ points ever since. The board continued to demand plans including redistricting and sought to compel compliance by withholding state financial assistance and obtaining further proposals from the Joint Center. The committee responded with lawsuits in the state courts to compel release of the withheld funds and annual bills in the state legislature, filed by committee members, to repeal the Racial Imbalance Act.19 In *440the rhetoric of the committee, redistricting became synonymous with abolition of the neighborhood school, anathema to most parents and to be resisted at all costs.
In other areas, the committee did submit plans. In 1967, it adopted a plan for ' the construction of new schools drawing pupils from more than one previously existing district; .for the reassignment of pupils in connection with the conversion of certain grammar schools to middle schools for grades 6, 7 and 8; an°d for the closing of three racially imbalanced schools and the reassignment, of their pupils. This plan was approved by the board “as a first step” and became known as the city’s First Stage Plan. Second and third stage plans in 1968 and 1969 provided for more new schools, but no redistricting. Meanwhile the number of racially imbalanced schools was increasing steadily, from 49 in 1966 to 62 in 1969.
The committee belatedly adopted a fourth stage plan on June 15, 1971, which was promptly rejected by the board on June 22. On July 12 the committee sent a letter containing other amendments to the plan, which the board rejected on July 20, stating,
“The letter contains no expression of intent to adopt a plan for redistrieting high schools and intermediate schools; it merely proposes further study. The letter also conditions high school redistricting on the development of a metropolitan plan.”
Under mounting pressure from the board and the courts, the committee voted further amendments on August 23, one for the revision of district lines and the other for comprehensive racial balancing, as follows,
“A Citizens’ Advisory Committee will be established to review the present district -lines for junior high, middle and high schools. Half of the members may be selected by the State Board of Education and half by the Boston School Committee. Staff members of both agencies would be available to assist in the effort. The Committee would present an initial report no later than January 15, 1972 and a Final Report no later than March 15, 1972. Implementation of new district lines, approved by the School Committee, designed to achieve racial balance in the junior high, middle and high schools shall be put into effect no later than September, 1972.”
“Comprehensive Plan
The process for preparing a Comprehensive Plan for racially balancing the Boston Public Schools will be undertaken as follows:
A. The School Committee will forthwith request technical assistance in preparing a Comprehensive Racial Imbalance Plan from the Commissioner of Education and the U.S. Department of Health, Education and Welfare. ,,
B. A representative Committee of distinguished citizens will be created during the 1971-1972 school year to oversee the planning process carried out by the staff of the Boston School Committee with the assistance of the staff of the State Board of Education. Members of the aforementioned Committee will be selected by the State Board of Education and the Boston School Committee.
C. A timetable for planning will be developed by the aforementioned committee to assure implementation of the Comprehensive Plan, approved by the Boston School Committee, no later than September, 1973.
D. The Plan shall include, among other things, the concept of metropolitanization.”
The committee then proceeded to sabotage both proposals. To membership on the Citizens’ Advisory Committee it appointed persons known to be vehemently opposed to redistricting and encouraged them to stalemate discussions with members who had been selected by the board. Regarding the broader plan, it engaged *441in a similar charade: on August 31, 1971, the committee chairman wrote to the board, stating in part,
“In view of the commitment of the Boston School Committee, by its vote of August 23, 1971, to develop a comprehensive plan which will assure an integrated education for all the children of Boston, I am requesting that the Board of Education provide technical assistance to the Boston School Committee in the preparation of alternative plans to eliminate minority-group isolation in the public schools for the consideration of the School Committee, under the provisions of Section 403 of Public Law 88-352. I am also requesting that the Board of Education provide assistance in the presentation, adoption, and planning for implementation of one of these plans.”
On December 21, the board wrote to the committee seeking access to pupil data cards and other information necessary if it was to provide the technical assistance requested in the chairman’s letter dated August 31. On January 26, 1972, a new chairman of the committee had been installed and he replied that technical assistance from the board would not be in order until after the Citizens’ Advisory Committee had completed its work, but that, “Should the need for technical assistance arise in the future, we will make a specific request for it.” The other committee, the “representative Committee of distinguished citizens to oversee the planning process”, was never even organized; members were not even appointed.
Ill
Feeder Patterns
At a meeting of the school committee on June 15, 1971, a letter from the state board dated May 28, 1971 enclosing suggested district lines for certain intermediate and high schools came up for discussion, as follows:
SUPERINTENDENT: They have been screeching for district lines for high schools.
CHAIRMAN: We have district lines for district high schools.
SUPERINTENDENT: We have feeders.
CHAIRMAN: South Boston High and Hyde Park High, for example.
MR. HAMBELTON (associate superintendent) : You don’t have geographic boundaries. You have feeder patterns.
SUPERINTENDENT: You can say that pupils from such and such a school district may go.
CHAIRMAN: The school district has geographical boundaries.
SUPERINTENDENT: That’s right. Indirectly the answer is yes.
A few minutes later, when another member started speaking about a high school’s “basic district”, the discussion went off the record, presumably to reinstruct the member in the esoteric subject of feeder patterns. Some of the confusion is doubtless due to the fact that high schools in identifiable residential sections of Boston, e.g., East Boston, Roslindale and Hyde Park, are thought of as neighborhood schools with natural districts and are usually referred to as “district” high schools. Technically, however, as the superintendent endeavored to explain, high schools do not have geographical districts of their own — they have feeder patterns.
Enrollment at high schools is determined by a combination of seat assignments, preferences and options collectively called feeder patterns. They are promulgated in February of each year in superintendent’s circulars which incorporate by reference the geographical boundaries of intermediate school districts. Since 1966 they have always been prepared by Associate Superintendent Meagher; for many previous years they had always been prepared by Associate Superintendent Tobin. Their operation can be understood only in the context of the various types of high schools in the city and the various types of intermediate schools whose graduates are qualified for high school admission, viz., *442junior highs, i.e., grades 7, 8 and 9; middle schools, i.e., grades 6, 7 and 8; and elementary schools ending in grade 8. Half of Boston’s eighteen high schools are “fed” by all intermediate schools throughout the city and are called “citywide” schools. They are the examination schools, Boys Latin, Girls Latin and Boston Technical; the trade schools, Boys Trade and Girls Trade; the high school component of the model demonstration subsystem, Copley Square High; a special ungraded school offering a work-study program, not mentioned elsewhere in this opinion, Boston High; and two general schools, English High for boys only and Girls High for girls only. The other nine, Brighton, Burke,20 Charlestown, Dorchester, East Boston, Hyde Park, Jamaica Plain, Roslindale and South Boston, are fed only by designated intermediate schools, with the qualification that under the open enrollment and controlled transfer policies, which permit transfers to fill vacant seats, pupils from anywhere in the city may theoretically attend any high school. This qualification must itself be modified because several “district” high schools have been overcrowded and transfers to them under either policy have been relatively rare. In this division of the opinion, the findings will pertain only to the general, i.e., non-speeialized, high schools offering the usual secondary school currieulm, viz., citywide English and Girls (henceforth in this division “Girls” will always mean Girls High and not Girls Trade) and the district high schools.
A further distinction between English and Girls and district high schools is that they include ninth grades, whereas six of the district schools, Brighton, Charlestown, Dorchester, Hyde Park, Jamaica Plain (apart from a small ninth grade for agricultural studies) and Roslindale, do not. These six contain only three grades — 10, 11 and 12 — and can be fed only by junior highs, all of which have grades 7, 8 and 9, or by transfers from citywide high schools. The three district high schools with grades 9-12 are East Boston, South Boston and Burke for girls, whose percentages of white students in September 1967 were, respectively, 99:%, 100% and 25%. The natural progression of students is obviously from intermediate schools ending in grade 8 to high schools starting with grade 9; and from intermediate schools ending in grade 9 to high schools starting with grade 10. It was one of several factors, together with overcrowding at some schools and the reluctance of blacks to attend virtually all-white schools and of whites to attend virtually all-black schools, which enabled the defendants, by the use of seat preferences and options, to keep the white and black races as separate as practicable at the high school level.
Since Associate Superintendent Meagher took over the responsibility for feeder patterns in 1966, they have been manipulated with segregative effect. Until then, circulars assigned high school seats only to pupils residing in designated junior high districts; pupils from elementary school districts ending in grade 8 either entered grade 9 at a junior high or enrolled at a citywide high school. In February 1967, on the basis of informal talks with the principals of some of the affected K-8 school districts but without consulting the headmaster at English, Mr. Meagher established feeder patterns for the ten K(indergarten)-8 grade elementary school districts, to become effective the following September at the start of the 1967-68 school year.21 These ten K-8 school districts are listed according to the percentages of their racial compo*443nents, with the whitest first; in the parentheses “W” signifies the percentage of white students at that school during the 1967-68 school year, “OM” signifies other-minority, and “B” black students; the schools under “Schools Fed” are high schools unless otherwise indicated :
K-8 Schools Schools Fed
W OM W OM B
Cheverus 100 E. Boston 99 1
Russell 95.8 3.5 S. Boston 100
McKay 91.8 8.2 E. Boston 99 1
Parkman 88.6 .3 11.1 Irving Jr. 93.6 1 6.4
Thompson 72.2 .3 27.5 Lewenberg Jr. 54.2 .5 45.3
Wilson Jr. 93.2 .3 6.5
Prince 50.6 11.4 38.0 Taft Jr. 77.6 4.2 18.2
Edison Jr. 78.5 1.5 20
Rlce43.1 22.5 34.4 Burke 24.7 8 66.3
Franklln * Girls 14.5 6.4 79
Lincoln 31.7 35.3 33.0 Burke 24.7 8 66.3
Girls 14.5 6.4 79
Martin * 31.3 2.2 66.5 Burke 24.7 8 66.3
Dearborn 11.4 .4 88.2 Burke 24.7 8 66.3
English 76 5.3 18.7
Girls 14.5 6.4 79
The effect of seat preferences at the five junior high schools listed in this table can be understood by showing the racial compositions during the school year 1967-68 of the high schools fed by these junior highs pursuant to preexisting feeder patterns. The high schools, all containing grades 10-12, at which they had seat preferences, and the years when first established are as follows:
Junior Highs Schools Fed
W 0M B
Irving 1958 Rosllndale 96.6 3.4
Lewenberg 1958 Dorchester 79.1 .3 20.6
1962 Hyde Park 96 .3 3.7
Wilson 1958 Dorchester 79.1 .3 20.6
1962 Hyde Park 96 .3 3.7
Taft 1962 Brighton 66.6 2.9 30.5
Edison 1962 Brighton 66.6 2.9 30.5
In analyzing these tables, the distinction between coeducational and non-coedueational schools is relevant, and also that between high schools with grades 9-12 and those with only grades 10-12. Thus it appears that students at five of the six majority white K-8 schools were fed to grade 10-12 high schools, and that the exception, Russell, was fed to South Boston, 100% white; and that students at the other schools, where whites were in the minority, were all fed to grade 9-12 type high schools. In the latter group of four K-8 schools, seat preferences were assigned only to girls graduating from three of them (since Burke and Girls high schools were for girls only), and to both boys and girls at the fourth, Dearborn. It should also be noted that the significance of a seat preference at a particular high school depends in large measure on whether the school is overcrowded or underutilized 22 — if the former, a seat preference would be a necessity for admission; if the latter, it would be of slight significance because vacant seats could be occupied without seat preferences by students transferring under the open enrollment policy which was in effect until August 1971, as described in the next division of the opinion. By this standard, preferences *444were significant in the school year 1967-68 only at three district high schools — Hyde Park where there was 95 more students enrolled than the school’s rated capacity, Roslindale with an excess of 121, and Brighton with an excess of 163 — and at citywide English which had a rated capacity of 1,200 students and was accommodating 2,103, an overenrollment of 903. Finally it should be noted that during the school year 1967-68, white students at English outnumbered black students by a ratio of 4:1.
A second set of feeder pattern changes was promulgated in February 1968 applicable to the school year 1968-69. Students at five K-8 schools were given additional seat preferences at English Burke or Girls. A sixth change pertained to Michelangelo junior high and to three schools whose grade structure was new to the local scene: middle schools containing grades 6-8. The idea of pupil progression from elementary schools with grades K-5 to intermediate schools with grades 6-8 to high schools with 9-12 had caught on in other parts of the country. Its supporters claimed various educational justifications for it, e. g., the earlier maturity of teenagers, the desirability of long-range career-planning, reducing the number of “drop-outs” on the occasion of changing schools, etc. Defendants introduced this new grade progression system in Boston beginning with the 1968-69 school year. In 1967 a new school, McCormack, had opened with grades 4-8 near the Columbia Point housing project and it was designated as a middle school. Two others, Lewis and Campbell (since renamed Martin Luther King), had been junior high schools23 and were converted to middle schools by adding grade 6 and eliminating grade 9. The fourth was Thompson, theretofore a K-8 school, whose overlapping district was shown on the first of the maps, ante, and which was included in the first table, ante, since its graduates were given seat preferences at two junior highs by feeder patterns promulgated for the' 1967-68 school year. There have been no additional conversions to middle schools or openings of new ones since 1968. A third category of change in feeder patterns in 1968 was the deletion of former seat preferences for graduates of two of the new middle schools, Lewis and Campbell, and two junior highs, Lewenberg and Wilson. The various feeder pattern changes at the different tpes of schools, together with the racial compositions of their students bodies for the school year 1968-69, are set forth in the following tables:
K-8 Schools Schools Fed
W OM B W 0M B
Russell 94.3 2.8 2.9 Burke 19.6 11.1 69.3
Girls 9.6 7.8 82.6
English * 66.1 9.7 24.2
English 34.6 8.9 56.5
English • 72.1 3.6 24.3
Prince 47.9 14.2 37.9 Burke (as above)
English
RiceFranklln 24.8 39.7 35.4 English
Lincoln 20.9 50.8 28.3 English
Martin 16 7.5 76.5 Girls
English
*445Intermediate Schools Schools Fed
W OM B W OM B
Michelangelo 96 4 Charlestown 98 .5 1.5
Thompson 65.9 31.1 Burke 19.6 11.1 69.3
Girls 9.6 7.8 82.6
English * 66.1 9.7 24.2
English 34.6 8.9 56.5
English 72.1 3.6 24.3
McCormack 27.6 66.4 Burke (as above)
Girls
English
S. Boston 98.8 .22
Campbell 4.6 94.8 Burke (as above)
Girls
English
Intermediate Schools Seat Preferences Discontinued At
W OM B W OM B
Wilson 91.5 .7 7.8 Hyde Park * 95.5 .6 3.9
Lewenberg 32.5 2.3 65.2 Dorchester * 72.3 1.5 26.5
Campbell .5 4.6 94.9 Dorchester (as above)
Lewis ** .5 99.5 Jamaica PI. 55.5 4.2 40.3
As before, in analyzing these tables the type of high school, i. e., whether 10-12 or 9-12 and coeducational, is relevant as well as the racial compositions of the K-8 and intermediate feeder schools and of the high schools fed; and also the preferences assigned a year earlier to all five K-8 schools. With respect to the tables showing intermediate schools, a further fact is essential to an assessment of the Michelangelo entry: since 1962 Michelangelo’s graduates had had seat preferences only at English. Also the preferences assigned to graduates of the middle schools must be appraised in view of the overcrowding at English24 and the preferences which were discontinued as well as the changing racial compositions of the feeder and fed schools. Finally it should be added that very few feeder pattern changes were made after those for 1968-69; and all changes of any consequence are described in this division of the opinion.
When the feeder pattern changes for the two school years 1967-68 and 1968-69 are considered together, it is apparent that: (a) graduates of heavily white K-8 schools — Cheverus, Russell and Parkman — were given seat preferences at heavily white coeducational high schools; (b) graduates of heavily black lower schools — -Martin, Dearborn and Campbell — were given seat prefer*446enees at English for boys and girls or Burke for girls and nowhere else; (c) graduates of lower schools having substantial white percentages who were given seat preferences at English, Girls and Burke — Prince, Thompson and McCormack — were also given options of going to predominantly white high schools either directly or indirectly by way of a predominantly white junior high; (d) when the feeder pattern changes applicable to English started to take effect, students at heavily white Michelangelo which had fed English for several years were given the option of attending heavily white Charlestown;25 and (e) channeling of black students to English, Girls and Burke and away from the grade 10-12 type high schools was facilitated by the selection of two of the most heavily black junior high schools in the city for conversion to middle schools.
Only once did the defendants make a seat preference change tending to alleviate racial imbalance; in 1971 McCormack middle school was deleted as a feeder for English, Girls High and Burke, leaving its graduates with a seat preference only at South Boston High. However, the deletion was more in the nature of a public relations gesture than a genuine change for the following reasons: (a) by September 1971 a seat preference at English had become meaningless due to the large number of vacant seats — between the 1967-68 school year, when the Meagher feeder patterns first took effect, and the school year 1971-72, total enrollment at English had plummeted from 2,103 students to 977 students; (b) English and Girls High received students on a citywide basis anyway; and (c) South Boston was 100% white, filled to capacity, and black students were believed to be unwelcome there in 1971. Predictably, the change in the feeder pattern made no difference: South Boston continued 100% white and black graduates of McCormack continued to enroll at English, Girls High and Burke.
Another feeder pattern change effective in 1971-72 was also of minor significance: K-8 Prince’s option to Taft junior high (but not to Edison junior high), which had been established in 1967, was eliminated, thereby narrowing the channel to Brighton for white students at Prince. However, in the interim enrollment at Prince had declined steadily from 324 students to 285 and the percentage of whites from 51% to 40%. The remaining seat preference at Edison was probably sufficient to accommodate the smaller number of white graduates from Prince.
The only consistent basis for the feeder pattern designations, changes and deletions was the racial factor. Neither distances between schools, capacities of receiving schools, means of transportation or natural boundaries explain them. In their requests for findings, defendants offered explanations for the seat preferences established for graduates of five of the affected schools — McCormack, Parkman, Prince, Michelangelo and Thompson — but their explanations have not withstood close scrutiny. The gesture regarding McCormack has been discussed, ante. Defendants submitted that Parkman and Prince had seat preferences prior to 1967-68; but the pertinent testimony was simply that in prior years some students at those K-8 schools enrolled at the Irving, Edison and Taft junior highs, not that seat preferences had existed. Defendants requested a finding that “the Prince preference at the Taft was eliminated because of space limitations at the Taft. The Edison has much more space available.” In fact, when the change was made, Taft was underutilized with a capacity of 540 and enrollment of 521 and Edison was overcrowded with a capacity of 786 and enrollment of 799; and a year earlier, when the Prince change was first promulgated, overcrowding at Edison was twice as severe as at Taft: an overenrollment of 56 students at Edison compared to 27 at Taft. Defendants sub*447xnitted that Michelangelo graduates were given the option of attending Charles-town in 1968-69 because the tenth grade for English was then located at Brighton; however, only 165 sophomores at English were sent to Brighton, the large majority of them remaining in English’s main building; moreover, the Charles-town option was undisturbed when all segments of the English tenth grade returned to the main building in 1969-70. Regarding Thompson, defendants’ requests are rebutted by evidence too complex to set forth in detail; suffice it to say that they overlooked the deletions of feeders from Lewenberg and Wilson junior highs, remaining options, and the fragmentation of the Lewenberg and Wilson districts in 1969-70 and 1971-72 into specified subdistricts.
With respect to the disproportionate impact on black students of the opening of the four middle schools and related feeder patterns, and the fact that no middle schools were opened after 1968, defendants have contended that the program of changing to a K-5, 6-8, 9-12 progression would have continued but for obstacles beyond their control which frustrated plans for a new 5,000 capacity campus type high school in Madison Park. However, defendants’ contention is inconsistent with the following evidence: (a) the only school committee member, present or former, who testified at the trial had no recollection of any connection between plans for a K-5, 6-8, 9-12 progression and the proposed campus high at Madison Park; (b) defendants’ 1966-67 racial imbalance plan discussed middle schools specifically but did not mention the campus high as part of that conversion; and (e) after the four middle schools were opened in 1968, a number of new K-6 schools were opened, rather than K-5 schools. which would have been compatible with a K-5, 6-8, 9-12 system. For example, the Holland K-6 school opened in 1972; it was in the King (formerly Campbell) middle school district which was 94% black; but Holland was predominantly white and its graduates were admitted at Cleveland junior high, 91% white, despite the fact that Cleveland had a capacity of 810 and an enrollment in 1971-72 of 1,131 and King had a capacity of 1,010 and an enrollment of 837, i. e., Cleveland was overenrolled by 321 pupils and King had 173 available seats. The explanation was given in the testimony of the associate director of the city’s Educational Planning Center: parents of pupils at Holland did not wish their children to go on to King. Finally, there was always abundant space available at 9-12 grade high schools which might have accommodated ninth graders if other junior highs than Campbell (renamed King) and Lewis had been converted to middle schools. Vacant seats at Girls between 1967 and 1972 averaged in excess of 500; at Burke there were 142 vacant seats in 1969-70, 185 in 1970-71 and 141 in 1971-72; English became underutilized, with 223 vacant seats in the main building in 1971-72. The problem, of course, was that the student bodies at those schools were predominantly black. This situation, however, did not exist in East Boston, where the high school contained grades 9-12 and there was a junior high, Barnes, nearby. The student bodies at both schools between 1967 and 1972 always exceeded 97% white. Moreover, East Boston high was always underutilized, with vacancies ranging from 212 in 1967-68 to 89 in 1971-72; and the ninth grade at Barnes probably had fewer than 250 pupils. If the defendants still intended a general conversion of junior highs to middle schools, the school in East Boston would seem to have been ripe for such a change, but was never converted. In sum, the lack of the large campus high school does not explain defendants’ failure to make further conversions to a middle school system. It is of course irrelevant as a justification for the discriminatory feeder patterns.
The consequence of the feeder pattern changes and discriminatory options, in combination with the opening of four middle schools, was altogether foresee*448able, almost immediate, and well-understood by the defendants: a dual system of secondary education was created, one for each race. Black students generally entered high school upon completion of the eighth grade, and white students upon completion of the ninth. High school education for black students was conducted by and large in citywide schools, and for white students in district schools. White students were generally given options enabling them to escape from predominantly black schools; 26 black students were generally without such options. The advantages and disadvantages of the introduction of middle schools were experienced almost entirely by one of the races, the black. The advantages and disadvantages of coeducation were experienced mainly by one of the races, the white. The high schools to which black students were channeled, especially English, changed in racial composition virtually overnight. During the first school year affected by the Meagher feeder patterns, 1967-68, black students at English accounted for 18.5% of the student body. In 1968-69, the entering class at English was 56.5% black and was segregated more than a mile away at Egleston Square. In 1969-70, the entering class at English was 76% black and 18.5% other-minority. During 1967-68, 1600 white students attended English; in September 1969, the number of white students enrolling as freshmen at English was 15. In the school year 1972-73 the student body at English as a whole was 81% black; and at Girls and Burke, both were over 90% black. Elsewhere, except at Dorchester where black enrollment rose from 20.6% in 1967-68 to 52% in 1971-72, increases in the percentages of black students in the student bodies of high schools were negligible.
The defendants knew what was happening. On October 15, 1970 the headmaster at English wrote to Mr. Meagher, who had become deputy superintendent, giving him the racial breakdown of the student body for the school year just begun, including 614 blacks and 542 whites, and listing four sources of the problem of increasing imbalance, as follows:
“1) the assignment of core-city feeder schools in our district;
2) the relatively few seats actually available to boys from these feeder schools when they apply elsewhere under the open-seat plan;
3) the lack of Grade 9 seats in two of our feeder schools and also in most alternative choices of high schools;
4) the tendency of white families to react to our trend toward minority imbalance by withdrawing their sons and thus accelerating the trend;”
At a school committee meeting on March 9, 1971, feeder patterns were a subject of discussion and the following exchange occurred:
MEMBER #1: So the fact is that the feeder schools are imbalancing the high schools, and the people won’t come in because they realize that?
SUPERINTENDENT: Therefore, the large city schools are becoming smaller and smaller and becoming black.
MEMBER #1: And they are underenrolled in comparision to the others?
SUPERINTENDENT: Yes
MEMBER # 2: On the questions of districts for high schools or feeder schools for high schools, that is what we face now. We have a choice. We can either at the English High School and Jeremiah Burke High School and Girls High School and perhaps Girls *449Trade School stay or become almost wholly Negro, the way they are now, or are becoming, or we can so redistrict as to say No, in which case we scatter those Negro youngsters into Charlestown, into East Boston, into South Boston.
SUPERINTENDENT: That’s right.
MEMBER #2: Do you want that degree of hullabaloo. How are they going to get into South Boston? If you are going to keep them out of those schools, how are they going to find room in those other high schools unless the other high schools are so redistricted to their feeder schools that they have to come to English High School?
SUPERINTENDENT: That’s right.
MEMBER # 2: People living in the Fort Point Channel area in East Boston could be told they must go to English High School and couldn’t go to East Boston High School
CHAIRMAN: I think we had better move along.
The committee’s initial fourth stage plan voted on June 15, 1971 contained the following statement:
“Traditionally large numbers of high school aged students from each residential district of Boston attended secondary school at in-town locations. However, in recent years this trend has been reversing. Numbers of secondary students have been abandoning the central schools in favor of district' schools, thereby contributing to the imbalance of selected high schools.”
Defendants’ implementation of these feeder patterns and discriminatory options, which was equivalent to the redistricting of several high schools, occurred at a time when they were rejecting and evading state board demands that they redistrict elementary and intermediate school districts to alleviate racial imbalance. At the lower levels, maintaining the status quo, e. g., neighborhood schools and open enrollment, often served to perpetuate segregation. At the secondary school level, however, where citywide enrollment was traditional at half of the city’s high schools and students routinely traveled several miles to and from school by public transportation, segregation could not be maintained without basic complex changes being made. The defendants made such changes for the purpose of promoting racial segregation and accomplished their purpose.
IV
Open Enrollment and Controlled Transfer
The terms open enrollment and controlled transfer refer to school committee policies which permit pupils, on an individual basis and when space is available, to enroll in schools outside of their residential districts or other than those determined by pertinent feeder patterns. In 1961, when open enrollment was first adopted, it was generally thought to be an aid to integration because it enabled black students to attend predominantly white schools if the schools had vacant seats and private transportation could be arranged. It soon became evident, however, that open enrollment was also an aid to segregation because it enabled white students to transfer from schools with racial compositions not to their liking. Under the Racial Imbalance Act, Mass.G.L. c. 71, §§ 37C and 37D, the state board tried to persuade and then compel the defendants to limit open enrollment by prohibiting transfers which increased racial imbalance. The defendants refused and for a period of about five years, 1966 to 1971, the issue was a hard bone of contention. Finally on August 23, 1971, as part of a package deal, i. e., one of the amendments to the defendants’ fourth state plan, whereby state funds which had been withheld since May 25 were released, defendants agreed to prohibit transfers which would increase racial imbalance. Plaintiffs have contended in *450these proceedings that the defendants’ change was more apparent than real.
The segregative consequences of defendants’ transfer policies, while by no means minimal, were small when compared to those caused by facilities utilization, districting and feeder patterns already discussed. Transfer applications are made and handled on an individual basis, whereas discrimination in facilities utilization, districting and the like affect large groups of students. Also, as will be explained, the extent of segregative transfers is blurred by loose use of the phrase “racial imbalance.” Nevertheless, the defendants’ words and actions in determining and administering their transfer policies are especially probative on the question of their intent.
Suggestions for changes in the open enrollment policy began in June 1963 when NAACP representatives addressed a school committee meeting on a broad range of subjects. In April 1965 the Kiernan Report suggested modifications of open enrollment, including (a) preference for transfers reducing racial imbalance, (b) free transportation if beyond walking distance, (c) centralization of administration instead of principals’ unsupervised discretion and (d) listings of vacant seats. In September 1965 a private group of black parents raised funds for private transportation of black pupils transferring under open enrollment to predominantly white schools. Called Operation Exodus and led by a black parent, Ellen Jackson, who testified at the trial, it bused approximately 250 black students in 1965-66 and grew steadily to approximately 1,100 in 1969-70.27 At some of the transferee schools, the students encountered locked doors, physical segregation in separate classrooms, auditoriums and corridors and placement in the rear of classrooms. Anticipating the arrival of black students, administrators of some transferee schools had desks unbolted from the floor and removed from classrooms.
A black parent, Underwood, and Operation Exodus complained in 1969 of racial discrimination in the administration of open enrollment to the Massachusetts Commission Against Discrimination (MCAD), which conducted hearings for twelve days and issued its findings on June 22, 1971. MCAD ex rel. Underwood v. Boston School Committee, No. EDXIV-l-C. The Commission concluded that open enrollment was administered with discrimination on the basis of race and color and issued a cease and desist order. It found, among other things, that (a) parents and children considering open enrollment application were given little information or help and were often misinformed or misled at the schools the children were already attending and by school department headquarters, (b) available seat-count figures were not distributed or made easily accessible to students seeking open enrollment admissions, (c) parents and students were discouraged from attempting to make applications under open enrollment by principals, administrators and guidance personnel, (d) personal interviews were used to determine the race or color of the applicants and to discourage applicants not desired by the receiving school, (e) the seat-count of available places was not based on established capacity figures and was a subjective compilation under the direction of the particular school principal not reflecting an accurate count, and (f) open enrollment was a means of effecting the transfer of students arbitrarily selected.
When the defendants refused to comply with the Commission’s orders, enforcement proceedings were instituted in October 1971 in the Superior Court, Equity No. 94218, which, on April 22, 1974, affirmed the MCAD findings but remanded for a determination of mootness since the complaining student had graduated and open enrollment had been superseded by the controlled transfer policy. Specifically the court found that open en*451rollment had been operated in a discriminatory manner so as to hinder and in many eases exclude black and other minority students from transferring to schools that had empty seats. On May 28, 1974 an MCAD commissioner reported that (a) the proceedings are not moot, (b) under the controlled transfer system, many of the practices previously found and condemned continue to exist, and (c) the school committee has not eliminated the discrimination that existed in the open enrollment system.
For a brief, ten-day period in June 1966, the open enrollment policy was amended to prohibit transfers aggravating racial imbalance. The background was as follows: in February 1966 the board rejected the defendants’ 1965 racial imbalance plan and made several proposals including changes in open enrollment. On April 12, 1966 the board voted to withhold state funds from the Boston schools. At a committee meeting on June 13, the chairman moved that the open enrollment policy be amended as follows: no pupil may be permitted to attend a school out of district if his attendance in that school will either cause said’ school to become racially imbalanced or aggravate an already existing condition of racial imbalance in such school. In support of his motion, the chairman stated,
If years ago we recognized that there was merit to the proposition [that a segregated education is educationally harmful] and took steps to withhold this privilege of open enrollment, today I don’t believe the Patrick T. Campbell would be an imbalanced school. I don’t think that the Solomon Lewenberg would be approaching an imbalanced state.
In other words, we could have avoided the difficulties that we now face. And I think it’s reasonable to predict that if we don’t adopt some kind of order such as this we are going to have more imbalanced schools creating more problems to be attacking in the future.
This seems to me to be the kind of preventive step we should take now while there is still an opportunity to take preventive action.
A dissenting member responded,
Mr. Chairman, I think it is wrong to take away this right, and I don’t call it a privilege. It is absolutely a right that has been given to the people of the City of Boston in the form of open seat enrollment.
The motion passed 3-2. However, the state board would not retract its disapproval of the defendants’ 1965 plan. Therefore at the committee’s next meeting on June 22, one of the June 13 majority moved to rescind, explaining that the state funds had not been released and also that he now believed that the amendment limiting transfers was an unconstitutional discrimination against white students. The latter point was endorsed by the superintendent who said that he believed that “our open enrollment policy is the finest open enrollment policy of any school system in America.” 28 Rescission was voted by 3-2. Every year thereafter until 1971, similar amendments to open enrollment were proposed and rejected.
The basis of the defendants’ resistance to curtailment of open enrollment was crystal clear: it would deprive white students of their “right to escape” from schools with substantial percentages of black students. By 1970 the member who had switched his vote at the June 22, 1966 meeting had been elected by his colleagues as chairman of the school committee. On the same issue at a January 20, 1970 meeting, he stated the same position he and other members had often explained at previous meetings, as follows,
*452Of course the thing that would have everyone deeply concerned would be again a school like the Lewenberg which starts to become evenly balanced or has become so and the white youngsters start to apply under open enrollment to move out, and under this they would be pretty much chained to their seats. They wouldn’t be allowed to move.
The defendants’ ultimate decision in August 1971 to adopt a controlled transfer policy was made upon consideration of a staff memorandum dated July 9, 1971 outlining available options and describing open enrollment as “parental choice, as to school attendance, historically granted to families in changing neighborhoods.” (Emphasis added.)
Open enrollment did not die without a struggle. On October 27, 1970, the state board voted unanimously to modify the open enrollment policy as follows:
1. Children may enroll in a school other than that of their assignment,' on a space-grade available basis, if, and only if, such enrollment decreases racial imbalance (as defined in the Act) in his school of regular assignment, and/or in the selected school;
2. Exceptions to the foregoing policy may be made only with respect to schools that offer specialized courses or curricula that are not otherwise available to eligible students ; individual exceptions to the foregoing policy may be granted by the School Department upon the basis of a written showing of hardship by the parents of the student-applicant;
3. The foregoing modifications shall be operative for the school year 1971-72, and to that end the Title IV, Equal Educational Opportunities Bureau of this Department is directed to provide forthwith to local School Committees and Departments such technical assistance as may be necessary to ensure implementation of this requirement.
At the committee’s next meeting, defendants voted to invite State Commissioner Sullivan to attend a meeting of the committee on November 23 “so that he can try to justify his position.” The Commissioner appeared and gave a comprehensive justification of the board’s action, stating in part that it was based upon the unanimous conclusion of the courts that no plan which involves parental or student choice of schools is permissible where it results in racial separation of white and Negro pupils; and that the Constitution does not permit school officials to facilitate racial separation of pupils. The defendants’ response was a vote at a meeting the next day to sue the state board and seek a declaratory judgment that the board’s vote of October 27 was itself unconstitutional. Suit was never instituted. On March 31, 1971, the commissioner requested assurances from the committee by April 16 that open enrollment would be eliminated. When such assurances were not forthcoming, the board on May 25 notified the appropriate state officers to withhold funds otherwise payable to the Boston school department. On June 15 the defendants submitted their fourth stage plan to the board without any modification of the open enrollment policy. On June 16 the board extended its May 25 withholding of funds and on June 22 it rejected the fourth stage plan. On July 12 the committee amended its plan but left the open enrollment policy intact. On July 20 the board rejected the amendments. On August 16 the committee amended its plan a second time, again leaving open enrollment intact. On August 17 the board voted a resolution requiring that open enrollment be eliminated. Finally, on August 23, the committee yielded, or at least seemed to.
During 1970-71, over 6,700 pupils were attending out-of-district schools under open enrollment, and at least 1,028 of these transfers were segrega*453tive: 324 non-white pupils transferred from majority white to majority nonwhite schools and 704 white pupils transferred from majority non-white to majority white schools. Beyond that, however, the extent of racial segregation caused by open enrollment in 1970-71 is impossible to pinpoint. During the same year, 1,546 white students transferred from one majority white school to another; and 1,660 black students transferred from one majority black school to another. But how many of these transfers were segregative cannot be determined on the basis of the evidence received at the trial. Part of the difficulty lies in the definition of racial imbalance in § 37D of the Act: when the per cent of non-white students in any public school is in excess of fifty per cent of the total number of students in such school.29 Thus racial imbalance is quite different from racial segregation. Suppose, for example, a two-school district with an equal number of white and black grammar school students in which all black pupils are assigned to school A and all white pupils are assigned to school B. Both schools would be segregated but only school A would be imbalanced. Yet often the terms were used interchangeably, as illustrated by the following discussion at a committee meeting on October 19, 1971 regarding proposed expansion of a secondary school facility:
Chairman: The important thing is that they need an addition out there at South Boston High. Who is going to fill it is a decision that is going to be made later. But we don’t want to stop progress as far as planning for the addition is concerned. Do you follow me ?
Member # 1: I follow you, but to be realistic about it, I don’t think the people in South Boston are going to want to have a racially balanced school.
Member # 2: The people in South Boston have a racially balanced school.
Chairman: To go forward on the addition to South Boston High.
Member #1: It would have to be racially balanced.
Chairman: Racially balanced could mean a hundred per cent white.
Member #1: I don’t think it would under Commissioner Sullivan.
Member # 2: Of course not.
Member # 1: Mr. Sullivan is committed to the numbers game, black and white — no question in my mind about that.
Mr. Hambelton (associate superintendent) : If we discussed this with the state and we do not show them how the school can be balanced or integrated, they will not fund it and that will end the project right then and there.
Although there was relatively little direct evidence as to the segregative effects of open enrollment, they were cumulative and probably substantial. Transfers under open enrollment were virtually permanent; they did not have to be renewed annually and they qualified transferred students for the feeder patterns applicable to graduates of the transferee schools. At school committee meetings during the period from 1966 to 1971, several members spoke of Lewenberg junior high as having become predominantly black because of open enrollment ; and individual members made the same statement about other schools, e. g., Holmes and Campbell junior highs and Prince elementary school. In sum, open enrollment as administered by the defendants became a device for separating the races and contributed significantly to the establishment of a dual school system.
The segregative effects of open enrollment were by no means remedied by the *454new controlled transfer policy because of its numerous exceptions, the most significant of which was a “grandfather clause” whereby “any student who last year attended an out-of-district school under the Open Enrollment Policy may continue to attend such school, so long as a seat remains available.” At the trial defendants sought to defend the grandfather clause on the grounds that it helped perpetuate transfers promoting integration and avoided disrupting friendships that had formed among students. The first ground is specious since transfers promoting racial balance would have been allowed anyway under the controlled transfer policy. As for preserving friendships, the record does not enable the court to discern why the defendants thought this social interest so important. There was no showing that pupils with friends in their out-of-district schools could not have made new friends in their in-district schools or that many of their friends would not also have been returned with them to the schools whence they came. A major consequence of the grandfather clause was to restrict severely the balancing potential of the new policy because nearly all otherwise available seats out-of-district were already occupied by transferees under open enrollment. The total number of transfers at all levels of schools under the new policy in 1971-72 was only 319.
The second exception to the controlled transfer policy came in the form of an interpretation of the grandfather clause by the superintendent with the approval of the school committee. Under date of August 23, 1971 the superintendent issued an explanatory circular including what were called “other considerations”, as follows:
1. All students presently attending out-of-district schools under the former Open Enrollment Policy may continue to do so.
2. No new transfers may be effected under the discontinued Open Enrollment Policy.
3. Those pupils who avail themselves of their options under this Controlled Transfer Policy beginning in September 1971 shall be required to re-apply for their seat on an annual basis.
Continuation shall be denied only in the case where increased attendance by pupils living within the district shall have produced a situation of extreme overcrowding.
4. Special educational needs and other “hardship” cases should be decided on a case-to-case basis by the Assistant Superintendent of the home district.
The fine point lay in the second consideration which embodied the superintendent’s view that the grandfather exception applied not only to students attending out-of-district schools under open enrollment during 1970-71 but also students who in May 1971 had filed applications for transfers which would begin during September 1971. Such transfers would not be deemed “new” because applied for before the termination of the open enrollment policy. This construction of the grandfather clause was, to say the least, difficult to understand, so during the committee meeting on September 7, 1971 the defendants went into executive session in order that the superintendent might explain the situation in private. After his initial explanation, the superintendent said, “I would hope that my circular would stand. I don’t think anybody will notice it but ourselves.” When the associate superintendent said later during the executive session that he would agree with the superintendent’s interpretation “if the State understands it the same way”, the superintendent replied, “To my point of view, the State isn’t going to find out about it until it’s too late.” Standing alone, the superintendent’s cynical comments might be discounted due to the strain and pressure of the moment— school was scheduled to open the next day, parental furore over the districting for the new Lee school was rising, etc. *455But their significance may not be so minimized. They must be considered together with similar remarks by committee members at previous meetings and their frequent expressions of disdain for the state board of education and state officials responsible for enforcing the Racial Imbalance Act. In that context the superintendent’s remarks accurately reflected the bad faith which characterized many of the school committee’s dealings with state officials.
A third exception to the controlled transfer policy covered transfers between schools in multi-school elementary districts. As seen ante, several elementary districts contain more than one school and they vary greatly in racial composition, e. g., the Marshall district, in which the Marshall and Champlain schools are within a half mile of each other and the student body of the former, in 1971-72, was 59% white and the latter 94% black. The board complained that this exception did not correspond to its understanding of the new policy which provided that, “Before October 15, a student can only transfer from a school in which his race is in a majority to a school in which his race is in a minority.” (Emphasis added.) The superintendent replied that the board should have understood all along that there would be such an exception. It remained in effect and allowed intra-district segregative transfers to continue.
A fourth exception may be called, for lack of a better term, “group exceptions” voted by the school committee. One, already described, permitted 200 black students to remain in Lee school although they did not reside within its district. Constituting approximately one-fourth of Lee school’s population, this contributed heavily to Lee’s opening with a 74% black enrollment. A related, extraordinary group exception — never voted, however, by the school committee — flowed from the option given in September 1971 to students encompassed by the new Lee district who had been attending field, 82% white, and O’Hearn, 99% white, schools: although residents of the new Lee district, they were permitted to continue in attendance at Fifield and O’Hearn. At the trial it was established that the same option has been continuously granted to the original beneficiaries’ younger brothers and sisters as they have entered the school system.30 Another group exception allowed 14 white Chittick graduates to avoid their district junior high, Lewenberg, which is 95% black. The asserted basis was safety because of a high crime rate in the neighborhood, but the same crime rate had to be endured by black students attending Lewenberg.
A fifth exception to the controlled transfer policy was “hardship transfers”, presumably granted only on “the basis of a written showing of hardship by the parents of the student-applicant.” However, hardship transfers were regarded and administered merely as a loophole and were explicitly referred to by school committee members as an “escape clause” and “a big out.” Sometimes these transfers were granted on explicit racial grounds and at others without any showing of hardship at all. The total number of hardship exceptions was small, 118 during 1971-72, yet hardly de minimis when compared to the total number of racially balancing transfers during the same period, 319.
Viewed together, the open enrollment and controlled transfer policies were managed under the direction of the defendants with a singular intention to discriminate on the basis of race. For open enrollment and evasion of controlled transfer restrictions were antithetical to the defendants’ foremost publicized policy to have each pupil attend the school serving his neighborhood *456community.31 The court has already found that the “neighborhood school” policy was no impediment to segregative districting, redistricting, use of facilities and feeder patterns. So here, when a neighborhood started to change from predominantly white to black, the “neighborhood school” policy was subordinated to the white students’ presumed right to escape to safely white out-of-district schools. The result of the defendants’ maneuvering was to encourage and facilitate the abandonment by white students and parents of schools which appeared to be in the process of becoming predominantly non-white.
V
Faculty and Staff
Plaintiffs’ claims in this area, unlike those heretofore discussed, concern policies of the defendants which discriminated against them indirectly— but no less significant!y.32 Another distinction is that the racial composition of neighborhoods, segregation in housing, distances to schools and safety precautions are irrelevant here. See, e.g., Kelly v. Guinn, 9 Cir. 1972, 456 F.2d 100, 107. Except for substitutes called in on short notice, teachers are not assigned on the basis of residence. Plaintiffs’ principal claim involving faculty and staff is that the defendants have knowingly pursued policies resulting in racial segregation of teachers and administrative personnel, thereby reinforcing the racial identifiability of schools and increasing the racial segregation of students. Plaintiffs’ next claim under this heading raises a further distinction in that it does not depend on the Brown and Keyes and other school cases decided by the Supreme Court in the past two decades, but on traditional equal protection principles. See Sweatt v. Painter, 1950, 339 U.S. 629, 70 S.Ct. 848, 94 L. Ed. 1114. Specifically, plaintiffs assert a denial of equal educational opportunity, viz., that defendants have knowingly pursued policies resulting in less qualified, less experienced and lower paid teachers in predominantly black schools. A third claim is that defendants have violated the constitutional right of public school students to have the school system operated free of racial discrimination in the recruiting, employment and promotion of teachers and staff. See Johnson v. San Francisco Unified School District, N.D.Cal.1971, 339 F.Supp. 1315, 1332. We find that plaintiffs have established all three claims.
A general introductory description of certain aspects of teacher employment, classification and transfer will help to clarify discussion of this group of plaintiffs’ claims. Probably typically, the Boston school system has always been structured in a way which would impede participation by outsiders generally. The vast majority of teachers have been products of local colleges and universities, especially Boston State Teachers College. New teachers were appointed from eligibility lists compiled on the basis of the Boston Teachers Examination (BTE), a special essay type test given only in Boston. The only means of qualifying for nearly all administrative positions was to serve as a teacher in Boston for several years. As thought to be required by Massachusetts *457law, Mass.G.L. c. 151B, § 4(3), the system was colorblind, i.e., there was no reference to race on application, transfer or promotion forms or in a teacher’s file. In recent years many changes have been made. The teachers union, Local 66, AFT-AFL-CIO, and the school committee entered into a collective bargaining agreement in 1966. The BTE was replaced mainly in 1968 and entirely in 1970 by the National Teachers Examination (NTE) prepared by the Educational Testing Service (ETS) of Princeton, New Jersey. Several programs were started as a result of Title I of the federal Elementary and Secondary Education Act of 1965 (ESEA), 20 U.S.C. §§ 241a-244, including the model demonstration subsystem, elementary enrichment, and work-study. Sixty black recruits for provisional teaching positions were trained in a program federally funded under the Education Professions Development Act, 20 U.S.C. § 1091 et seq. However, organization and practices of the system remain much the same, e.g., the complex provisions in the union contract covering teacher transfers, at appendix C thereof, set forth procedures and criteria employed since 1951.
Three categories of teachers are employed in Boston, permanent, provisional and temporary substitute. Mass.G.L. c. 71, § 38G provides,
“No person shall be eligible for employment by a school committee as a teacher, . . . unless he has been granted a certificate by the board; . and provided, further, that a school committee may upon its request be exempt from the requirements of this section by the department for any one school year when compliance therewith would in the opinion of the department constitute a great hardship in the securing of teachers for the schools of a town. This section shall not apply to trade, vocational, [and] temporary substitute teachers. . ."
Permanent teachers have obtained certificates from the state board which requires a bachelor’s degree from an accredited college and courses of study and semester hours therein as established by the board. Permanent teachers have various rights under state statutes and the union contract.
Provisional teachers are full-time teachers who have college degrees but lack state certificates, presumably because they haven’t taken courses in education prescribed by the board. Also, they may not have taken the NTE examination or have done poorly on it. They receive one-year contracts pursuant to the exemption in § 38G for which Boston evidently - qualifies each year by showing “a great hardship in the securing of teachers”, although there is an oversupply of applicants for permanent positions to such an extent that on the major eligible lists the ratio of teacher applicants to positions is approximately 10 to 1. Approximately 700 or nearly 15% of the city’s full-time teachers are provisionals. They are hired to fill vacancies which are not permanent and when there are no candidates on a particular sublist of the eligible list and, beginning in 1971, for purposes of minority recruitment. A provisional teacher may become a so-called permanent provisional by serving for three consecutive years, after which he acquires tenure under Mass.G.L. c. 71, § 41. Provisionals are compensated at a substantially lower rate than teachers with certificates: in the first year of employment, $1500 less and in the third year, $2100 less. Provisionals have no rights of transfer or promotion.
The third category of teachers are temporary substitutes who fill in for ill or absent teachers on a day-to-day basis and whose compensation is at a rate lower than that of provisionals. There are also a large number of practice teachers still in college and teachers’ aides of one sort or another, usually hired under federal programs.
Vacancies in teaching positions and in building level administrative positions are filled first by transfers within the system on the basis of seniority. The *458seniority criterion has been employed in effecting transfers since 1951 and since 1966 has been included in the collective bargaining agreement between the defendants and the teachers union. The necessary conditions for a transfer are that a faculty member be a permanent teacher and have taught in that capacity for at least the two previous years in a Boston public school. Given an opening for which he or she is qualified and sufficient seniority, a permanent teacher may transfer every two years. The procedure is as follows: by means of superintendendent’s circulars issued twice a year, permanent teachers are notified of the conditions of their right to transfer and of the availability of request forms; there are spaces on the transfer request form for five choices of school districts or sections of the city to which the applicant wishes to be transferred; the forms are filed with the personnel department which compiles lists of requested transfers, ranking the requesting teachers according to seniority which is measured by years of permanent teaching employment in the Boston public school system from date of employment; when vacancies in permanent positions are about to occur, principals and headmasters notify the personnel department on Request for Permanent Teacher forms and the vacancies are filled on the basis of seniority from the transfer lists. A teacher maj1 refuse an offered transfer which he has requested, but if he does so he may not transfer until the following year.
Transfers are made only in accordance with the procedure described. Technically the right of permanent teachers to transfer is subject to denial'for the good of the system, but a transfer has evidently never been denied on that ground, at least never on the ground that it would increase racial imbalance. The superintendent also has power to make transfers for the good of the service; the following provision is included in every individual teacher employment contract :
“It is understood that the School Committee of the City of Boston reserves the right to transfer or assign a teacher from one school, district, department, position, grade or subject to another as the interests of the Boston Public Schools demand.”
But this power has never been exercised.
If a vacancy in a permanent position is not filled by a transferring teacher, it is normally filled by hiring a certified teacher from an eligible list of applicants who have successfully finished the screening process for hiring new permanent teachers. Applicants are listed in the order of the score they achieved in the screening process, which takes into account grades on the NTE examination, an interview and credentials such as graduate study and experience. Assignments to particular schools are frequently made on the basis of requests by headmasters and principals. Occasionally, when no certified applicant is available, permanent positions are filled by provisional teachers. At elementary and intermediate levels, transfers and appointments vare made to districts, not schools, permitting the principal of a multi-school elementary district to assign teachers to schools within his district at his discretion.
Vacancies in building level administrative positions not filled by. transfers, and central administrative positions, are filled by promotions. Boston’s promotion system was suspended in the spring of 1971 and vacancies since then have been filled on an acting basis. Before 1971 from four to six years of experience were required to qualify for promotion. Candidates for promotion applied to the board of examiners, which issued promotion certificates on the basis of written examinations and interviews. Persons qualifying for promotion were placed on a “rated list” compiled for each type of position, e.g., headmaster, assistant principal, etc. Such lists were drawn up only every third year. Administrative positions of assistant su*459perintendent and above are not filled by rating, but by the superintendent’s selection and school committee approval.
Racial Segregation
Black teachers are segregated in black schools. In 1972-73 there were majority black enrollments at 59 of the city’s 201 schools. Of the total of 356 black teachers, permanent and provisional, 244 were stationed at those 59 schools. The percentages of the total number of black teachers in the system who were teaching at majority black schools during the five previous years follow:
1967-68 67%
68- 69 68%
69- 70 70%
70- 71 72%
71- 72 74%
A rough understatement of the situation is that less than one-third of the schools are majority black, but over two-thirds of the black teachers are sent to them.
Whatever the angle from which a view is taken, the picture is the same. In 1971-7233 the defendants hired 82 black teachers, permanent and provisional, for the first time. Of this total approximately three-quarters, precisely 61, were assigned to teach at that fraction of the city’s schools, less than one-third, which were majority black. Exactly half of the new black teachers were assigned to seven schools whose enrollments were over 90% black: Champlain, Higginson, Holmes, King, Lewenberg, Paine and Timilty. In 1971-72 there were 158 elementary schools in the system and 167 black teachers were assigned to them. However, three or more black teachers were assigned to the following 18 schools, whose total enrollments and numbers of white pupils and percentages of white pupils of the total enrollments are also listed:
Elementary Schools Black White Total White % Teachers Students Students of Total
Bacon * 2 199 1%
Baker 9 466 2
Dearborn * 5 456 1
Dlckerman * 0 311 0
Ellis 6 600 1
Emerson 59 333 18
Fenwick 12 317 4
Garrison 4 904 .4
Gibson 25 589 4
Greenwood 13 800 1.6
Hale 5 286 1.7
Howe * 4 422 1
Hurley 23 464 4.9
Lee 229 1089 21
Marshall 699 1185 59
Mason 72 261 28
Paine 25 795 3
Perkins 1 125 .8
Lee and Marshall might well be eliminated from the table, since Lee was so special a situation and Marshall is a new so-called magnet school. Of the remaining 16 schools to which three or more black teachers were assigned, only Emerson and Mason had student enrollments as high as 5% white. Fourteen elementary schools with three or more black teachers on their faculties had student enrollments in which white students comprised less than 5%. In nine such schools there were fewer than ten white pupils in the entire school. In four (marked with an asterisk), white pupils were outnumbered by black teachers. On the other hand, as of 1972-73 no black classroom teacher, permanent or provisional, had ever been assigned to 81 of Boston’s 201 schools; and an additional 35 schools have had only one black teacher in any year since 1967-68, the earliest year for which figures were put in evidence.
Black administrators are also segregated. In 1972-73 all five black principals were at majority black schools, as follows: Emerson, 66% black; Lewis, 72%; Tileston, 72%; King, 95%; and Timilty, 97%. There were fourteen black assistant principals and assistant headmasters and all were assigned to *460majority black schools, as follows: Trotter, 53% black; English, 81%; Burke, 91%; Champlain, 93%; and Timilty, 97%. Black temporary teachers are also segregated, partly because a proper factor in their selection is proximity of the teacher’s residence to the school needing a day-to-day substitute on short notice. Another reason, however, is that index cards of blacks desiring substitute work often include the names of either of two minority recruiters and this is used by the personnel department to match the substitute’s race with the requesting school’s racial composition. Figures as to substitutes were not put in evidence.
Racial segregation of full-time faculty and staff has been attributable first to requests by principals and headmasters that black teachers be assigned to their districts. Such racial requests are received from ten' to fifteen district principals a year and are honored by the personnel department whenever possible. Senior school administrators and the defendants were well aware of this practice. The second cause was the tendency of permanent teachers who had taught at majority black schools for at least two years to exercise their transfer prerogative to transfer to majority white schools. In 1971 and 1972, a total of 100 such requests were made and 39 were granted. In the same period, only two requests for transfers from majority white to majority black schools were made, and neither was granted. In the same period there was but one transfer from an elementary school with a lower percent of black pupils to an elementary school with a higher percent — it was from a school with a black enrollment of 50-60% to one with 80-90%. There was no educational justification for these transfers, i. e., there was no particular need for more experienced teachers in the receiving schools. These facts are related to the practice of filling permanent vacancies by transfer before new appointments and the current oversupply of teachers. Candidates who are not near the top of the eligible list may find that their only chance to enter the system is to start teaching in a predominantly black school; then after two years they will request a transfer. This situation is well known to senior administrators and the defendants; assistant superintendent Griffith, testifying on deposition about a meeting of the board of superintendents, said,
“That there should be some policy of retaining teachers. Because, as we see it, it is easier for a person to gain access into the school system by coming in through a troubled school. If you look at the top of a list and you have very, very high scores — I’ll not say nine out of ten times, but a goodly degree of these people will refuse appointments to troubled schools. So the teacher that wants to get in the system will take the troubled school, knowing that in two years they can move out.”
The pattern of faculty transfer from black to white schools in Boston was mentioned in the study report of the federal Civil Rights Commission filed in 1966; and the subject was discussed at several school committee meetings. Lastly, over one-third of the black teachers, 125 of 356, are provisional teachers with no transfer rights; and black permanent teachers generally have less seniority than most white permanent teachers. The transfer system’s exclusive reliance on seniority thus effectively locks many black teachers in the schools to which they are initially assigned, usually majority black, and allows white teachers with seniority to transfer from these same schools to others.
Defendants have argued that assignment of black teachers to predominantly black schools is educationally justifiable because black teachers serve as adult role models and inspirational examples to black pupils. The record, however, is barren of evidence supporting the argument. One witness, associate superintendent Leftwich, testified that a need by black youngsters for adult role models would in his opinion be a logical explanation for granting requests of black teachers for assignments to predominantly black schools. But his was concededly a curbstone opinion unsupported *461by any study or empirical data. The defendants never made any systematic inquiry concerning whether assigning black teachers to black schools inspires the confidence, or coincides with the wishes of, the black community. More importantly, there was no evidence that such considerations in fact played any part in the frequent matching of the races of teachers and pupils.
Defendants also contended that racial matching, when requested by black teachers, would have educational value because such teachers would likely be especially dedicated to the training and progress of black pupils. The first difficulty with this contention is the paucity of supporting evidence. The only evidence in the entire record that black teachers requested assignments to black schools had to do with two groups of black provisional recruits trained under the Education Profession Development Act (EPDA), 20 during the summer of 1971 and 40 in the summer of 1972. Assistant superintendent Griffith, who is in charge of minority recruitment, testified on deposition that the 1972 group of 40 “by and large preferred placement in black schools” and that the personnel department attempted to go along with their thinking and place them in the black schools. But there was no evidence' that the black recruits actually requested such assignments. Surely the legislative intent of Congress in passing the federal Act was not to segregate black teachers in black schools. The defendants have never inquired of black teachers generally about the kind or degree of their commitment, if any, to teach in black schools. More fundamentally, even if defendants had done so, and black teachers requested assignments to black schools, this would not overcome the prohibition of the Fourteenth Amendment which bans intentional racial segregation in public schools whatever may be the desires of black teachers or parents. Cooper v. Aaron, 1958, 358 U.S. 1, 78 S.Ct. 1401, 3 L.Ed.2d 5; Booker v. Special School District No. 1, D.Minn.1972, 351 F.Supp. 799, 809.
Unequal Education
The gist of plaintiffs’ claim here is that the defendants’ practices heretofore described result in predominantly black schools being staffed with less qualified and experienced teachers and with everchanging faculties, all to the detriment of black pupils who generally are receiving an education unequal to that being given white pupils. This claim is not geared to the racial composition of the faculty and staff at predominantly black schools but to the quality of education available to students attending them. The facts support plaintiffs’ claim.
Uncertified, provisional teachers, of whom there were 585 in the system in 1971-72 and 690 in 1972-73, are found in much greater numbers in predominantly black schools than in predominantly white schools. Statistics were compiled for 1970-71 and 1971-72 and a clear pattern with but minor exceptions emerges in both years: the blacker the school, the larger the percent of provisional teachers. The following table shows the distribution of provisional teachers as related to the racial composition of elementary schools in 1971-72 but is representative of percentages at schools at all levels in that year and in 1970-71:
% Black Students Number of Provisionals % Permanent Teachers
90-100% 24 schools 106 78.1%
80-90% 9 schools 34 74.6%
70-80% 5 schools 15 84.8%
60-70% 4 schools 19 85.7%
50-60% 4 schools 12 85.9%
40-50% 6 schools 25 79.2%
30-40% 9 schools 31 81.2%
20-30% 4 schools 14 82.9%
10-20% 14 schools 19 91.0%
0-10% 78 schools 60 94.6%
*462Note that schools above 80% black enrollment had an average of more than four provisional teachers per school; and that schools under 20% black enrollment had an average of less than one provisional teacher per school. A similar pattern emerges when one calculates the years of teachers’ service in the Boston school system in relation to the racial composition of the schools to which they are assigned. This was done for elementary schools in 1970-71, with the following results:
% Black Students Average Years in Boston Schools
90-100% 5.67
80- 90% 5.02
70- 80% 7.26
60- 70% 6.40
50- 60% 7.74
40- 50% 8.95
30- 40% 8.52
20- 30% 11.33
10- 20% 10.20
0- 10% 13.53
This table is consistent with the previous one; it also suggests that the system may channel teachers into progressively whiter schools as their seniority increases until they arrive and remain at schools less than 30% black.
Provisional teachers are not merely novices, but novices whose lack of qualifications and training necessitate their exemption from the provision of state law, Mass.G.L. c. 71, § 38G, that “No person shall be eligible for employment by a school committee as a teacher unless he has been granted a certificate by the board.” They may be given one-year contracts without even having taken the NTE or other standard examination which tests their subject matter competence. Of course it is possible, as argued by defense counsel, that a particular provisional may be a better teacher than one who has taken the usual courses in education and has received a creditable score on the NTE and when interviewed has compared favorably with other applicants for appointment. But the odds are against it. Generally, it cannot be doubted that the professional qualifications and skills of provisionals are inferior to those of permanent appointees. They are paid lower salaries, have slim prospects of permanent appointment and no prospects of promotion or transfer and generally lack the stature of newly appointed permanent teachers.
As in any calling, experienced practitioners are ordinarily more effective than novices, regardless of certification. The expert witness for the defendants so testified. A highly capable principal who was selected to head Lee school, Miss Frances Kelly, testified,
“I wouldn’t want to see a school, any school, with too high a percentage of first-year teachers in it, because the first-year teacher does need helping hands from experienced ones around him and if you have too many first-year teachers in a building, percentage-wise, then the experienced teachers can’t keep up with their errors and the thing becomes very unmanageable.”
The need for more experienced teachers in black schools was discussed at school committee meetings on several occasions; sometimes it was referred to as an “experience imbalance.” At a meeting on March 10, 1970, a member stated,
“These are the schools where we should have teachers who have much experience. Our most experienced teachers are needed there. The fact of the matter is that the inner city schools have a far greater proportion of first year, second year, and third year teachers.”
Problems attributable to teacher inexperience in black schools were known to senior administrators as well as to the defendants.
The combination of a disproportionate number of provisional teachers at black schools and the right of permanent teachers to transfer after serving at a school for two years visited another inequality upon black students, viz., a rate of teacher turnover higher than at predominantly white schools, referred to sometimes as staff instability. This subject was discussed at several school *463committee meetings. On March 28, 1972, for example, the chairman of the parents’ advisory board at Lewenberg junior high, then 95% black, addressed the committee regarding the poor performance of students there on citywide tests, as follows,
“Our main concern is the low level of student output at Lewenberg Junior High School. Our children exhibit low skill levels in arithmetic and reading.
The first issue I shall speak on is the issue of staffing. Our position is that you take action to stabilize the staff. Today most staff members at the school are new to it and to teaching. In fact, many members of the staff are provisional teachers, and, therefore, have no chance to become permanently appointed to the building.”
The defendants and senior administrators were aware that the high rate of teacher turnover at predominantly black schools was educationally harmful. Associate superintendent Leftwich testified, “The high rate of turnover, there is no question in my mind, is a bad thing and needs to be addressed in terms of remedy.”
Therefore the court finds, as requested by plaintiffs, that teacher resources in Boston have been disproportionately allocated on a systematic basis and this distribution operates to deny black students an equal educational opportunity. Hobson v. Hansen, D.D.C.1971, 327 F.Supp. 844; Spangler v. Pasadena City Board of Education, C.D.Cal.1970, 311 F.Supp. 501, 514-515, 524; and Johnson v. San Francisco Unified School District, supra, 339 F.Supp. at 1318, where the court stated,
“The law is settled that school authorities violate the Constitution by assigning . . . teachers of limited experience to ‘black’ schools while assigning few, if any, such teachers to ‘white’ schools.”
Hiring and Promotion
For 1972-73, there were 4243 permanent teachers in the Boston school system, of whom 231, or 5.4%, were black. A state professional staff survey for 1970-71, the only year as to which such a survey was offered in evidence, gave a racial breakdown of the various administrative positions in the Boston school system for that school year, as follows:
Position Whites Blacks
Principal and Headmaster 76 3
Assistant Principal and Assistant Headmaster 194 11
Director of Department and Manager 27 0
Assistant Director of Department and Assistant Manager 49 22
Supervisor and Consultant 29 2
Attendance Supervisor (Truant Officer) 46 0
Psychologists 28 0
Teachers Assigned to Home Bound Instruction 42 0
Total 491 18
Blacks filled approximately 3.5% of the positions. These percentages may be contrasted with those of black students in the system, approximately 33%, and of the city’s black population, approximately 16%.34 Plaintiffs contend that the disparity is due to racially discriminatory hiring and promotion practices knowingly followed by the defendants, and hence violative of plaintiffs’ rights under the Fourteenth Amendment.
Plaintiffs’ complaint concerning hiring practices rests largely upon defendants’ misuse of the National Teachers Examination (NTE). It is based upon legal principles discussed comprehensive*464ly in Castro v. Beecher, 1 Cir. 1972, 459 F.2d 725. Permanent teachers in Boston are selected from eligible lists and offered positions in the order of their ranking, which is determined by scores comprised of three components: the results of an examination, a personal interview, and credentials such as previous teaching ratings and years of experience. Until 1968 all applicants took an essay examination prepared by the school system’s board of examiners. Known as the Boston Teachers Examination (BTE), the examination was never formally validated, i. e., determined to be significantly related to the capacity of applicants to become good teachers. From 1968 until 1970 applicants were allowed to take instead the Teaching Area Examination of the National Teacher Examinations (NTE), which is prepared by the Educational Testing Service (ETS) of Princeton, New Jersey, and administered throughout the country. In 1970 the BTE was discontinued. At the time of the switch the board of examiners, which is responsible for hiring teachers in Boston, set a cut-off score of 560 out of an accepted maximum of 800 points on the NTE, a figure that matched the 70% grade formerly required to pass the BTE.35 Persons who “pass” the NTE by scoring above the cut-off are accorded brief interviews, which are held in Boston; the interviewer marks applicants in four equal areas: appearance, use of English, personality, and attitude toward teaching. In practice, an applicant’s total rank score (NTE, interview, credentials) has an almost perfect correlation with his NTE rank score, i. e., nearly everyone receives the same total score on the interview and credentials parts of the grading. Thus, as a practical matter, whether and in what order applicants for permanent teaching positions are hired depend almost entirely on their performance on the NTE.
Defendants’ use of the NTE in this fashion conflicts with the recommendations of Educational Testing Service which prepares it. ETS has emphasized in several publications that the NTE results should not be the determining criterion for who is hired and who is not. The reasons are that the NTE has not been shown to have substantial predictive validity, i. e., it measures only a fraction of the characteristics required for effective classroom performance; and its use assumes, in the words of an expert witness, “a kind of precision to testing that just doesn’t exist.” Studies conducted by ETS, of which Boston school officials are aware, indicate that blacks tend to score more poorly on the examinations than whites. Evidence of racial differences in success on the NTE is generally known to professionals in education. The effect of the cut-off 36 is to eliminate relatively more black applicants than white, and of the ranking by NTE score, to reduce the number of blacks employed as permanent teachers. Discriminatory use of the NTE has been expressly ruled to be unconstitutional. Baker v. Columbus Municipal Separate School District, N.D.Miss.1971, 329 F. Supp. 706, 713-717.
I Keenly aware of the strikingly small proportion of black teachers and administrators in the Boston school system, the defendants have for years “gone through the motions” of recruiting black teachers, but have never made a wholehearted effort to get results. The burden has fallen mainly on the shoulders *465of the highest ranking black official in the Boston school system, Rollins Griffith, who, as an assistant superintendent, has charge of the schools in one of six geographical areas of the city. Mr. Griffith’s area embraces most of Dorchester and part of Roxbury; it has the highest percentage of black students of any area in the city. Mr. Griffith entered the system as a music teacher in 1951 and has risen steadily. In 1966 he founded the Massachusetts Negro Educators Association, now named Black Educators Association of Massachusetts (BEAM), and was its first president. In 1969, a coalition of black parents urged that he be named principal of the Campbell (now King) school and he was promoted to that position, although never “rated” for it. In 1970 he was named assistant superintendent. For several years he has been the school systern's overworked “resident expert” on matters involving black students, parents and teachers and problems of concern to the black community in general. Mr. Griffith, with one assistant, has made several trips to colleges and conventions in an effort to interest black teachers in coming to Boston. The results of his efforts have been, in his own - words, “very, very poor”;37 and the reason has been obvious enough: lack of/ time, money, authority and cooperation: from the defendants. Mr. Griffith has never been authorized to offer permanent employment “on the spot” to black candidates whom he has interviewed at distant colleges and universities, as have recruiters from other cities, including Newton, Massachusetts. He and his assistant have been allowed only to recruit black teachers willing to enter the Boston system at the provisional level. The budget for minority recruiting is much less than in other large cities and was reduced from $60,000 for 1970-71 to $34,000 for 1971-72.
Except for accepting the federal EPDA program already described, resulting in the employment of 60 black provisional teachers (of whom at least eight had left the system by the time of trial), the defendants’ contribution to minority recruitment has been much talk but little action. At a meeting on July 31, 1972, a proposal to recruit black truant officers was approved and later tabled — at the time of trial there were no black truant officers, now known as supervisors of attendance, in Boston. At three other committee meetings in 1972, Mr. Griffith sought committee approval of plans to increase the number of black teachers, and none of his proposals was adopted. The defendants’ attitude is fairly reflected by the following exchange at a committee meeting on August 23, 1972:
Mr. Griffith: Yes. I am saying that we should take a position that we should say X number of minority teachers should be hired.
Member #1: That’s right, and we should tell white teachers that we can’t use them.
Mr. Griffith: Well, somewhere over the years some minority teachers have been told that they couldn’t be hired.
Member #1: No question about it. But sometimes two wrongs don’t make a right.
Until 1971, building level and central administrators were selected by means of a procedure known as “rating.” Only permanent teachers and administrators who had been employed in Boston in their current positions from four to six years qualified for promotion. Lists of personnel, “rated” for the various types of administrative positions 38 were com*466piled every three years and each list had a life of three years. The effect of this system was quite drastic: for example, a person appointed to a principalship in 1969 would have to have rated for the 1967 principal list, which would still be in effect in 1969, and would have had to have been teaching in Boston since the 1962-63 school year. A person wishing to “rate” applied to the board of examiners. A general prerequisite was a promotional certificate for the position desired, obtained by scoring 70% or better on an essay exam given by the examiners. This score was only a qualifying condition and did not enter the rating procedure itself, which was analogous to the initial hiring procedure: there were 800 possible points, 600 for credentials and 200 based upon an interview as to which there were no objective guidelines. A candidate needed 760 points or more to be rated.
Few blacks successfully completed the promotional process. Between 1969 and 1971 the board compiled 39 rated lists of candidates for promotion. Twenty-five lists contained no black candidates; eight contained only one black candidate; two contained two black candidates; and three contained four black candidates. The number of candidates who qualified for promotion during those three years was as follows:
White Black
1969 331 13
1970 115 7
1971 173 5
Because blacks comprised only an estimated 1% of the teaching faculty in Boston during the early 1960’s, rating requirements have had a continuing discriminatory effect, even as the number of black teachers has increased.
The school committee has shown its awareness of the problem of discrimination in the promotional process. In 1968 it commissioned a study of the promotional system by the local deans of schools of education; the resulting Deans’ Report was submitted in 1970 and recommended an overhaul of the system.39 However, the defendants acted slowly in responding to the Deans’ Report. The old system is apparently suspended' and the board of superintendents has been asked to recommend changes. The consequences of that system, however, remain: blacks still occupy a disproportionately small portion of Boston’s administrative positions. The school committee froze the promotional system in August 1971. Nonetheless the thirty vacancies filled at the time of trial on an acting basis had all been filled by whites.
VI
Examination and Vocational Schools and Programs
A high degree of racial segregation also exists in the city’s specialized high schools and vocational programs. Three high schools, Boston Latin, Girls Latin and Boston Technical, have traditionally admitted students only by competitive examination. They offer excellent college preparatory instruction and their graduates matriculate at the nation’s finest colleges and universities; they are called examination schools or elite schools. Their enrollments are heavily white; and the faculties at the three schools from 1967 to 1972 averaged only one black member per school. The precise percentages of black students at these high schools for the school years beginning in the fall of the years 1967 to 1972 were as follows:
1967 1968 1969 1970 1971 1972
Boston Latin 3.2 3.0 2.6 2.3 1.9 2.2
Girls Latin 3.5 6.1 6.3 5.3 5.4 5.0
Boston Tech.* 6.7 8.5 8.1 8.3 10.8 13.7
The two trade schools, Boston Trade and Girls Trade, have heavily black enroll*467ments. The precise percentages of black students in their enrollments for the period were as follows:
1967 1968 1969 1970 1971 1972
Boston Trade 9.6 40.7 41.3 53.2 66.5 65.2 *
Girls Trade 35.7 40.0 66.0 69.8 74.8 68.7 *
The city also operates various industrial trade programs at several regular high schools whose enrollments are predominantly white. The question presented is whether the racial segregation in these high schools and programs is intentional and therefore unconstitutional. The test to be applied is stated in Keyes v. School District No. 1, Denver, Colorado, supra, 413 U.S. at 208. On the basis of the presumption created by the defendants’ segregative practices in other parts of the school system, we conclude that these schools and programs have been intentionally segregated.
The controlling principles appear in the Keyes case which held that “a finding of intentionally segregative school board actions in a meaningful portion of a school system creates a presumption that other segregated schooling within the system is not adventitious. It establishes ... 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 segregative actions.” We have heretofore made detailed findings of intentional segregation of students and staff at all grade levels and in all parts of the city. Plaintiffs have shown racial segregation in the examination and vocational schools and programs, thereby establishing “a prima facie case of unlawful segregative design on the part of school authorities” with respect to such schools. The burden of disproving unlawful intent thereupon falls upon the defendants, who in this case have failed to carry that burden. The presumption created by their general course of conduct stands.
There was not a great deal of evidence received at the trial on the question whether the defendants intended that the examination schools be segregated; but most of what was introduced tended to support the validity of the presumption discussed in the Keyes case. Their racial identifiability is unrelated to the neighborhoods in which they are located: Boston Tech is surrounded by black residential neighborhoods, Girls Latin is located on the border of the black residential area in Dorchester, and Boston Latin is in the Fenway section, more of an institutional than residential area. Their enrollments are all citywide, with approximately 24 buses transporting students to each school. Admission may be at the level of the 7th grade or the 9th. The examinations given prior to 1973 were achievement tests for English and mathematics rather than aptitude tests. It is generally accepted that blacks perform poorly on this type of examination. Racial statistics as to the percentages of whites and non-whites who passed the examinations are available for 1971 and they confirm the generally accepted understanding, showing a widely disparate success rate for whites versus non-whites:
Boston Latin Girls Latin Boston Tech.
7th 9th 7th 9th 7th 9th
White 25.4 7.5 24.4 2.5 52.2 32.8
Non-white 4.3 0 6.6 0 21.5 9.2
While there is no specific evidence in the record that the examinations discriminate against blacks, computer printouts prepared by the defendants as early as 1966 demonstrated that students at schools with high percentages of whites had a success rate two to three times greater than students from schools with high percentages of blacks. In 1971 charges of racial discrimination in admission to the elite schools were filed with the Massachusetts Commission Against Discrimination. The defend*468ants agreed to a settlement of those charges in May 1972 which provided that for three years beginning in 1973 the defendants would use entrance examinations prepared by the Educational Testing Service of Princeton, New Jersey, and reserve seats, 84 at the Latin schools and 28 at Boston Technical for four students from each of the underrepresented districts of the city.
In addition to the full range of vocational training given at the citywide trade schools and the printing course at Boston Technical, whose racial compositions have already been shown,40 part-time cooperative vocational programs are offered at six district high schools 41 at which the percentages of black students from 1967 to 1972 were as follows:.
1967 1968 1969 1970 1971 1972
South Boston 0 0 0 0 0 .05%
Charlestown 1.3% 1.4% .8% 3.3% 2.0% 1.9
East Boston 1.1 1.9 2.2 1.6 2.5 4.4
Brighton 30.4 36.2 24.9 26.5 23.1 16.8
Hyde Park 3.7 3.9 7.8 12.0 15.2 18.8
Dorchester 20.6 26.1 35.4 37.9 52.1 57.7
These figures do not show the degree of segregation in the vocational programs as such, but do warrant an inference at least as to courses at South Boston, Charlestown and East Boston that enrollments in the vocational programs at those schools have been predominantly white. Also, in a few instances precise racial information about particualr vocational programs at some schools was received in evidence: (a) the electrical program at Charlestown had 406 students, of whom approximately 6 were black and 6 other-minority, (b) the machine shop program at Hyde Park had 94 students in grades 10 to 12, of whom 5 were black, and (c) of an unspecified number of students in the printing program at Boston Technical, about 20% were black.
The type of vocational training offered at the predominantly white district high schools is quite different from that at predominantly black Boston Trade. In the latter, students spend all their time at school and learn their trades in workshops at school. In the district high programs, the school conducts “cooperative-industrial” courses featuring part-time employment obtained by the school for vocational students in the 11th and 12th grades, which enables them to earn an average of $1,800 per year. Students may keep their jobs after graduation. Typically, a student alternates weeks in regular classes and on the job, e. g., a student in auto mechanics works every other week in a garage, where his performance is graded by his employer and reported to the school. There are educational advantages to both types of programs.42 On the one hand, students at Boston Trade receive more supervision by instructors and a broader range of experience than students gaining on-the-job experience in cooperative programs. A student at Trade may change from one *469vocation to another, whereas a vocational student in a district high school may not change without also changing schools. On the other hand, students in the cooperative courses may receive experience in some types of work not undertaken at Boston Trade because of the cost of materials. Regarding strictly academic courses, there are no differences other than the scheduling of classes. Students in the district high programs are not awarded a diploma, however, until they have completed one year of full-time work in their chosen fields after / their regular high school years.43 This withholding is wholly unrelated to academic studies since only the extra year of work experience is required for the award of the diploma; apparently it is designed to encourage students to remain in the same jobs after completing their classroom studies and to help them adjust to the transition from part-time to full-time employment.
As in the case of the examination schools, little evidence was offered on the issue whether the racial segregation of vocational students was purposeful; and the presumption of unlawfulness described in the Keyes decision was not rebutted. In fact, some independent evidence buttresses the court’s reliance on the presumption in finding unconstitutional segregation in the vocational programs: (a) there are no black instructors in the vocational programs at South Boston, Charlestown and East Boston— whether there are any in other district high programs did not appear in the evidence; (b) on-the-job training could be arranged for students at Boston Trade, but has not been; (c) cooperative-industrial courses have never been arranged for students at English, Girls High and Burke, the only predominantly black high schools other than the two trade schools; and (d) program coordinators of the district high programs focus their efforts to interest junior high school students in enrolling in eooperative-industrial courses on predominantly white junior highs.
Boston Trade and Girls Trade will both close when a new vocational facility opens, the Occupational Resource Center, which will be a central facility offering vocational programs to students at high schools throughout the city. Boston Trade opened in 1912 and has been in urgent need of repairs and renovation. In 1970 the man later selected to be director of the proposed new Center wrote a memorandum of reasons for not building a new four-year residential trade school; his first two reasons were:
“1. Statistics indicate that an alarming rate of pupils are dropping out of our trade schools.
2. Residential trade school pupils suffer from a sense of inferiority and the students going into it are generally regarded as second-rate people who cannot survive in the college-directed program.”
Residential Segregation and Neighborhood Schools
In addition to defendants’ contentions discussed ante, defendants have raised two major defenses which pertain to a number of plaintiffs’ claims. First, they contend that the segregation in Boston’s schools is the inevitable consequence of segregated housing patterns and the increase in the city’s black population for which they are not responsible. Second, they assert that they have adhered to a neighborhood school policy which is constitutionally valid regardless of any segregative consequences. These two arguments are not wholly separate; rather, they have both a common factual basis in this case and an analytical inter-relationship. For a school to reflect the racial composition of the area in which it is located, attendance must obviously be based upon a district which is to some degree the same as the area around the school. This “district” may *470or may not be a “neighborhood” and the school may or may not be a neighborhood school. However, the major thrust of the defendants’ first point would be that if the school district boundary predated a segregated housing pattern within that district, there is no constitutional violation. By contrast, the defendants’ second point about their alleged neighborhood school policy interjects something else: nonr acial, presumptively valid governmental reasons for decision making. These would include such things as the safety of the child by providing a nearby school and fiscal economy by avoiding the costs of transportation. Because the two defenses overlap, we shall discuss them together.
Initially it should be pointed out that these defenses are only partial, i. e., neither argument has any relevance to the defendants’ practices with respect to faculty and staff, open enrollment and controlled transfer or feeder patterns. These three areas of the defendants’ conduct, in which abundant evidence of segregative intent was found, are unrelated to residential segregation or neighborhood schools. Boston’s faculty and staff are not assigned on the basis of residence; open enrollment and controlled transfer are both inimical to a neighborhood school system; and high schools are not neighborhood schools.
There are various answers to the defendants’ arguments: first, it is now generally recognized that school officials’ practices may have a substantial impact upon housing patterns; second, when school officials have followed for at least a decade a persistent course of conduct which intentionally incorporated residential segregation into the system’s schools, that conduct is unconstitutional; and, third, when school districting and a neighborhood school system are fraught with segregative exceptions, neither defense need even be considered.
It is now generally agreed that schools and neighborhoods have a reciprocal effect upon one another. There is a substantial interaction: a school will cause the racial composition of the neighborhood to shift and vice versa. Both plaintiffs’ and defendants’ expert witnesses testified as to this. These phenomena have been recognized and discussed in many cases, e. g., Swann v. Board of Education, 402 U.S. 1, 20-21:
“People gravitate toward school facilities, just as schools are located in response to the needs of people.- The location of schools may thus influence the patterns of residential development of a metropolitan area and have important impact on composition of inner-city neighborhoods.”
See also Felice v. Board of Education of New York City, E.D.N.Y.1974; Higgins v. Board of Education of Grand Rapids, W.D.Mich., July 18, 1973, No. 6386. We shall not dwell upon the point beyond stating that in this respect Boston is not unlike other major urban areas, and that defendants may not disclaim responsibility for segregated schools because of population shifts which they themselves may have contributed to bringing about. Because of the complexity of the point and the difficulty of identifying and separating causes versus effects and their reciprocal dimensions, we rely more directly upon the considerations which follow.
The defendants have, with awareness of the racial segregation of Boston's neighborhoods, deliberately incorporated that segregation into the school system. This has been accomplished through construction practices and facilities utilization. With respect to school construction, the defendants have pursued a pattern of building relatively small schools to serve defined racial groups. These practices were not founded solely upon private discrimination, but also incorporated unconstitutional discrimination by other governmental officials. The 1960 United States census reported that 97% of Boston’s black population lived within Roxbury, Dorchester and the South End. This segregated housing configuration *471had not been altered noticeably by 1970 — only the number of black residents and the basic area in which they reside had expanded. The concentration in these three areas remained the same.
It has been asserted that residential segregation exists in our cities regardless of law, policies and other forms of discrimination. Taeuber, Negroes in Cities 36 (1965). We shall never know whether Boston’s segregated housing patterns would have evolved as they are today absent longstanding and thorough governmental complicity. Local, state and federal agencies historically followed discriminatory policies with respect to housing with consequences that remain clearly visible.44 Nor is there any present prospect of a reversal of the segregated housing patterns which were brought about by governmental action. The Boston Housing Authority built and maintained racially segregated housing projects, making tenant assignments on an explicitly racial basis. The Federal Housing Authority maintained an official policy of racial compatibility in housing. See 1938 FHA Underwriting Manual § 980(3)(g). As recently as 1971 Boston banks providing FHA mortgages as part of the Boston Banks Urban Renewal Group (BBURG) would grant mortgages to blacks only within a specific geographical area of the city. This was done for the avowed purpose of preserving racially homogeneous neighborhoods. These funds were virtually the only source of home financing for low and moderate income black families. The Massachusetts Legislature outlawed housing discrimination in 1950 with little if any result. The complementary consequence of private discrimination in the sale and rental of housing was also established by testimony at the trial. The combined effect of this private discrimination and the BBURG program was to limit blacks to a block-by-block expansion of the areas in which they could obtain housing during the late 1960’s. This kind of expansion, most commonly associated with the term “block-busting”, does not result in integrated housing.
In 1972 there were twenty-five housing projects in Boston; twelve were 79% or more white and thirteen were 73% or more black. This residential segregation was readily observable in Boston and widely known. In Boston the term “Roxbury” probably carries many of the connotations which “Harlem” does in New York City. Unfortunately, streets in disrepair, burned out houses left standing, and rubbish left on side streets in parts of Roxbury cannot’ help but intensify the identification of parts of this section of the city. The defendants were not ignorant of these segregated housing patterns and projects, which were a frequent topic of discussion at school committee meetings from at least 1963 until the time of trial. In fact the defendants had unusually detailed knowledge of existing residential segregation and forecasts for changes -because of a study commissioned by them in 1962. This study, the Sargent Report, predicted with 95% accuracy the size of the black population and those neighborhoods in Boston which would become predominantly black by 1970. Those predictions were confirmed by the 1970 United States census. The existence of this report is a sufficient answer to defendants’ intimation that they were surprised by shifting racial concentrations which frustrated their racial balance plans.
A substantial correlation exists between racial segregation in Boston public housing and in nearby schools. In 1972, for nine public housing projects which are 80% black, there were 33 schools, all of which are majority nonwhite.45 Similarly, 38 schools serving *472eleven 80% white housing projects are all majority white, and 29 of them are more than 80% white. This correlation between residential segregation and school segregation is the direct result of school construction projects between 1953 and 1972. Specifically, the small size of these schools was such that their enrollments could not but reflect the racial composition of the immediate neighborhood. Their locations were such that it is readily inferable that their racial compositions were intended; many were located within housing projects, bordered on three sides by them, directly adjacent to them, across the street from them, or on the same block. At least twenty such elementary schools were built near racially segregated housing projects; the Dever, Tobin, Hurley, Prescott, Carter, Aggasiz and Hennigan schools are prime examples. The defendants have also built additions to segregated schools or used portables which perpetuated the racial composition of those schools and neighborhoods. The following table demonstrates the extensive construction of school facilities near segregated housing projects and shows the racial compositions of both in the year when the new school, addition or portable opened: '
Year New Housing Percent White
Opened School Project School Project
1953 Fairmount Fairmount 100
1956 Guild-a Orient Hgts. 100
1957 Greenwood Fairmount 100
1957 Dever Columbia Pt. 93
1958 Bradley Orient Hgts. 99
1958 Grew Fairmount 100
1959 Tobin Mission Hill Ext. 25
1961 Hurley Lennox & Camden 2
1963 Prescott Charlestown 99
1964 Grew-a Fairmount 100
1964 Greenwood-a Fairmount 100
1967 McCormack Columbia Pt. 43 * 54
1967 Taylor-a Morton St. 83 98
1967 P. Bates-p Wash. & Beech 88 89
1967 Bigelow-p D Street 94 94
1967 Hoar-p D Street 91 94
1967 Palne-p Franklin Hill 69 79
1971 Carter Lennox & Camden 5 3
1971 Lee Franklin Field 21 13
1972 Agassiz South Street 81 79
1972 Hennigan Heath Street 24 1
"-a" signifies an addition to an existing school.
"p" signifies a portable.
Several of the schools and expansions shown in the table were opened as the racial composition of the nearby housing development was in the process of changing from white to black. Examples of such projects are listed as follows, together with the name of the nearby school or expansion and the percentages of the units in the projects occupied by white residents in the following years: (1) three years before the opening of the school or expansion, (2) the year in which the school or expansion was opened, and (3) in 1972, the latest year for which figures were introduced in evidence:
Housing School or
Project Expansion (1) (2) (3)
Mission Hill Ext. Tobin 38% 25% 4%
Columbia Pt. McCormack 74 54 27
Franklin Hill Paine-p 93 79 . 7
Franklin Field Lee 80 13 8
Heath Street Hennigan 24 1 1
These changes were due partly to the Boston Housing Authority’s practice, *473when black residents gained access to a predominantly white public housing project in more than token numbers, of avoiding referral of white applicants to that project but referring only black applicants.46 The extent to which these changes in the racial compositions of housing projects may also have been attributable to the reciprocal effect of a nearby school, alluded to ante at 121-122, is of course unknown. In any event, it should have come as no surprise to the defendants when the schools opened imbalanced and became more imbalanced thereafter. The opening racial composition of Tobin was not in evidence because it opened before 1967; it was shown, however, that in 1967-68 To-bin was 31% white. A possible exception to the pattern of imbalanced openings was the portable at Paine; the school as a whole, when the portable was opened, was balanced, i. e., 69% white. But by 1972, the school as a whole was heavily imbalanced with an enrollment of only 5% white, matching almost exactly the racial composition of nearby Franklin Hill project, by then 7% white. Conceding confusion as to some causal relationships, one consequence of the location of the new schools and expansions was clear: the containment of children, black and white, in schools near the housing projects in which they lived, effectively incorporating the segregation practiced with respect to public housing by other governmental officials.
The defendants have argued that they began a program of constructing larger elementary schools which could accommodate students from, a number of neighborhoods of differing racial compositions so that the schools would be racially balanced. It is true that larger schools were opéned, but not due to the defendants’ initiative. The defendants began to construct larger schools only in 1966 when the state refused to approve construction plans which would have resulted in more segregated schools. That year the defendants proposed the construction of three schools, one each in the Marshall, Mather and Gibson districts. Each of these schools, given their size and location as proposed by the defendants, would have opened racially segregated. The state board insisted that two larger elementary schools be built instead. Then when larger schools were opened, particularly Hennigan and Lee, defendants resorted to other kinds of segregative practices which have been the subject of previous findings.
As for the defendants’ asserted neighborhood school policy, findings heretofore made prove that it was so selective as hardly to have amounted to a policy at all. Several practices of the defendants were antithetical to a neighborhood school system: extensive busing, open enrollment, multi-school districts, magnet schools, citywide schools and feeder patterns. Additionally, the elementary district map, ante at 436, does not show districting which would be consistent with a neighborhood school policy: schools are not located near the center of regular, compact districts, but rather near the edges of irregular districts requiring some students to attend a relatively distant school when there is another school within one or two blocks. Notice, for example, the relative distances from a residence just east of the broken, dividing line to Emerson and Clap, Fenwick and Everett, and Dicker-man and Mather. When the defendants have emphasized considerations of distance and safety, it has usually been when they could be used to maintain segregation; students have seldom been the benefactors of this concern when there was no prospect of integration. The neighborhood school has been a reality only in areas of the city where residential segregation is firmly entrenched. Thus students in South Boston or Roxbury have been assured of attending a neighborhood school, all white *474or all black. White students who found themselves in predominantly black schools because of their residence were practically guaranteed a right to transfer under open enrollment or one of the exceptions to the controlled transfer policy • — regardless of overcrowded conditions at the receiving schools. Therefore, as the third response to the partial defense of residential segregation and neighborhood schools raised by the defendants, a further quotation from the Keyes case, 413 U.S. at 212, is apposite, as follows:
“[W]e have no occasion to consider in this case whether a ‘neighborhood school policy’ of itself will justify racial or ethnic concentrations in the absence of a finding that school authorities have committed acts constituting de jure segregation. . . . the mere assertion of such a policy is not dispositive where, as in this case, the school authorities have been found to have practiced de jure segregation . of the school system by techniques that indicate that the ‘neighborhood school’ concept has not been maintained free of manipulation.”
The arguments raised by the defendants in this case have been identical to those in the Keyes case in that they are mere assertions. See also Spangler v. Pasadena Board of Education, C.D.Cal.1970, 311 F.Supp. 501, 522; Davis v. Pontiac School District, E.D.Mich.1971, 309 F. Supp. 734, 744, aff’d 6 Cir. 1971, 443 F. 2d 573, cert. denied, 1971, 404 U.S. 913, 92 S.Ct. 233, 30 L.Ed,2d 186, and Taylor v. New Rochelle, 2 Cir. 1961, 294 F.2d 36, 39.
Inseparability of District
In considering the ramifications of the defendants’ segregative conduct, further attention must be paid to the city’s geography and the natural boundaries that lie within it. This necessity is explained in Keyes v. School District No. 1, Denver, Colo., supra, 413 U.S. at 200-205, from which the following excerpts are taken:
“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.” Id., at 200.
“This is not a case, however, where a statutory dual system has ever existed. Nevertheless, where plaintiffs prove that the school authorities have carried out a systematic program of segregation affecting a substantial portion of the students, schools, teachers and facilities within the school system, it is only common sense to conclude that there exists a predicate for a finding of the existence of a dual school system.” Id., at 201.
“In short, common sense dictates the conclusion that racially inspired school board actions have an impact beyond the particular schools that are the subjects of those actions. This is not to say, of course, that there can never be a case in which the geographical structure of, or the natural boundaries within, a school district may have the effect of dividing the district into separate, identifiable and unrelated units. Such a determination is essentially a question of fact to be resolved by the trial court in the first instance, but such cases must be rare. In the absence of such a determination, proof of state-imposed segregation in a substantial portion of the district will suffice to support a finding by the trial court of the existence of a dual system.” Id., at 203.
As previously mentioned, certain areas of the city, Charlestown, East Boston, Brighton and South Boston are fairly intact communities with somewhat restricted access to other parts of the city. The first two areas in particular are separated from the rest of Boston by the Boston inner harbor. Is either of them or any other part of Boston “a separate, identifiable and unrelated section of the school district that should be treated as isolated from the rest of the district”? Id., at 213. This inquiry was considered at the HEW administrative proceeding *475in which both the trial judge and the reviewing authorities adopted the government’s limitation of the Boston school system for the purpose of its analysis which excluded both East Boston and South Boston schools as being in “geographically distinct areas.” Initial Decision dated March 2, 1973, at 14. The administrative judge divided the Boston schools, for purposes of his opinion, into a central system, an East Boston subsystem and a South Boston subsystem. Id., at 72.
However, no section of the city is “separate, identifiable and unrelated” within the meaning of the Keyes case. Charlestown and East Boston are separate but in no way isolated. Ready access to Charlestown may be had by the Charlestown Bridge or the Northwest Expressway or the Charles River dam via Prison Point Bridge. East Boston and the center of the city are connected by two double-lane one-way tunnels, the Callahan Tunnel heading north and the Sumner Tunnel south. Each section is served by frequent public rapid transportation which connects with all other parts of the city and makes three stops in each section. Brighton47 is accessible both by public transportation and several major streets, including Commonwealth Avenue and Beacon Street.
Charlestown, East Boston and South Boston schools are in no way unrelated to those in the rest of the city. In making this evaluation several discussions, ante, are relevant but will not be repeated, viz., feeder patterns, overcrowding, open enrollment, vocational programs, and practices with respect to faculty and staff. All of the city’s schools are administered centrally by the school committee and the central administrative offices under the superintendent. For many years Boston’s schools were all citywide schools insofar as open enrollment was the official school policy; the only theoretical limits on this policy were available seats and the students’ desire to enroll.
Students from the three conceivably separate sections are transported on city-operated buses to the elite high schools located in other parts of the city, as shown in the following figures for 1972:
East Charlestown Boston South Boston
Boston Latin 38 60 84
Girls Latin 12 35 30
Boston Tech. 23 60 72
Total 73 155 186
Students from these same sections are also transported at city expense to Trotter and Lewis, the elementary and intermediate components of the model demonstration subsystem. Similarly, students from the other sections of Boston listed below were in attendance at public high schools in the three sections in question during 1972:
East South Charlestown Boston Boston
Roslindale 46 1
Roxbury 12 19
South Boston 20
West Roxbury 15
North End 105 2
South End 29 27 2
Charlestown 1 1
Dorchester 97 12 764
East Boston 18 1
Jamaica Plain 27
Total 369 48 788
Lesser numbers of out-of-district students also attended junior high schools in the sections in question.
Lastly, the defendants were parties to the HEW proceedings, Docket No. CR-982, 72-1, in which their main witness was William G. Tobin, deputy superintendent until his retirement in September 1970. The summary of his testimony, at 42 of the Initial Decision, states that Mr. Tobin “further asserted there was no area in the school system designated either as a central system, a South Boston system or an East Boston system. The entire system operated as a *476single integral system.” This court agrees.
State Defendants
The plaintiffs’ case against the state defendants, generally speaking, is that they have not exercised their powers to prevent various segregative acts by the city defendants. Plaintiffs point in particular to the fact that the state board approved racial balance plans that effectively promoted segregation. They also assert that the state defendants, the state board of education and commissioner of education, failed to reduce segregation by use of their specific statutory powers with respect to school construction, transportation aid, vocational programs and educational standards for professional personnel. In the court’s opinion, however, the evidence does not warrant a finding that the state defendants intentionally contributed to or participated in any substantial way in creating or maintaining racial segregation in the Boston public schools. On the contrary, the evidence established that they did everything within their limited authority under the law to compel the city defendants to obey the state law and the federal constitution.
True, the state board approved four separate racial balance plans for Boston. However, each plan, if it had been executed in good faith, might have made a substantial contribution to the elimination of racial segregation in Boston’s schools. The failure of these plans was uniformly the result of the city defendants’ failure to carry them out promptly and to take steps necessary for a particular proposal to have been effective. The state defendants consistently made or endorsed suggestions that the city defendants either refused to accept or accepted only after long delays, including redistricting, the use of portables to reduce segregation, changes in feeder patterns, restriction of open enrollment and transportation. Additionally, the state board voted several times to withhold financial aid because of what it considered to be failures on the part of the city defendants to comply with the state law or execute racial balance plans which had been approved. The withholding of state aid pursuant to Mass. G.L. c. 15, § II was the ultimate sanction that the board had under state law.
In at least two instances the state board forced the defendants to modify plans so that schools might open racially balanced. Marshall, which opened racially balanced, was originally planned by the city defendants on June 13, 1966 for only 700 pupils. The state defendants rejected that plan on June 28, 1966 and directed that the school be expanded to at least 1200 in order to accommodate several hundred additional white students. In February 1967 the city defendants agreed to expand the school plan to an enrollment of 1000 students and the state board approved. But for this expansion Marshall would have been majority black because of its location. Holland similarly opened racially balanced only because of the state board. The original plan of the city defendants on June 6, 1966 provided for two schools: one in the predominantly white Mather district and another in the predominantly black Gibson district. This plan was rejected by the state board, which recommended a single project combining those districts. The city’s February 1967 plan partially adopted this recommendation and the state contributed 65% of the cost.
The state defendants’ efforts to reverse the increasing segregation in Boston’s schools was considerably hampered by the limited authority conferred upon them by statute. The board was given only ultimate weapons, viz., the withholding of state funds and resort to the state courts, rather than authority over the management of Boston’s school system. Given the method of evasion adopted by the city defendants, formalistic compliance followed by procrastination and evasion on technical grounds, the means of enforcement at the board’s disposal were highly ineffective. Sometimes funds withheld to force the adoption of specific proposals had been re*477leased before the city defendants techniques for evasion became known to the board, as happened after the fourth stage plan was approved in August 1971. Similarly, judicial process is slow and unwieldy as a sánetion in the hands of an administrative agency operating under a statute as complex and technical as the Racial Imbalance Act. The state defendants were simply out maneuvered by the city defendants and frustrated by their intransigence and frequent bad faith. These attitudes were exhibited, for example, at a meeting of the school committee on April 8, 1971 at a time when $28,000,000 in state financial assistance was being withheld and the state board was requesting that the committee submit a fourth stage plan including districting of high schools and elimination of open enrollment. The committee was discussing how it should counter the state board’s demands. One member proposed a vote that the chairman of the board’s committee on racial imbalance be taken off the job. Referring to the city’s Education Planning Center which would draw up the city’s plan, the following exchange occurred:
Superintendent: My team will make your plan — not my plan but the School Committee’s plan. I have to keep saying that.
Chairman: We will be the bad guys. That is what we were elected for.
Eventually the state defendants charged unconstitutional segregation by the city defendants in the state court, where the claim was denied for insufficient proof, and in a crossclaim in this case. The racial imbalance plan which the state defendants finally succeeded in imposing upon Boston by Supreme Judicial Court orders dated January 16, March 22 and April 17, 1974, which is scheduled.to take effect in September 1974, is the latest demonstration of their best efforts.
If the court were concerned only about liability issues, judgment would be entered in favor of the state defendants; but, in view of its conclusions, it must also deal with the remedies to which plaintiffs are entitled. The state board and state commissioner have ultimate responsibility for the education of pupils in the Boston public school system. Therefore, although they have not contributed to the racial segregation that pervades the Boston schools, their continued presence in these proceedings may be necessary for the effectuation of appropriate remedies. The state defendants will be retained as parties defendant for purposes of formulating and aiding in the execution of such remedies. See Griffin v. County School Board of Prince Edward County, 1964, 377 U.S. 218, 234, 84 S.Ct. 1226, 12 L. Ed.2d 256; Rule 19(a), Fed.R.Civ.P.
Conclusions of Law
Keyes v. School District No. 1, Denver, Colo., 1973, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548, furnishes a blueprint for decision of the liability issues in this case, as follows: (a) whether racial segregation exists in parts of the Boston public school system; (b) if so, whether such segregation was intentionally caused and maintained by the defendants ; (c) whether any parts shown by plaintiffs to have been intentionally segregated constitute a substantial portion of the school system, i.e., whether defendants’ intentionally segregative conduct affected a substantial portion of the students, schools, teachers and facilities within the school system. If plaintiffs prevail on these issues, it follows that there is de jure segregation in the city and that the defendants’ obligation to eradicate it is no less than if it were due to a local ordinance requiring separate school facilities for different races. There will remain for decision, however, a fourth issue governing the nature and extent of the defendants’ obligation to desegregate and of the remedies which must be ordered by the court, viz., (d) whether, after considering any evidence that the system is geographically separable or that segregative intent was not among the factors that motivated the defendants’ actions, as a result of (a), (b) *478and (c) the entire school system is a dual system, one for white students and one for black students.
Plaintiffs proved beyond question that racial segregation exists in parts of the Boston public school system. That issue was virtually uncontested. In reaching this conclusion we have taken into consideration the racial composition of school enrollments and faculties and staff, community and administration attitudes toward the schools and factors of geography and demography. See Keyes v. School District No. 1, Denver, Colo., supra, at 196.
The paramount issue in this case is whether the defendants acted with a “purpose or intent to segregate” Keyes v. School District No. 1, Denver, Colo., supra, at 208, i.e., either with a desire to bring about or continue segregation in the Boston schools or with knowledge that such segregation was certain, or substantially certain, to result from their actions. See 1 Restatement of Torts 2d, § 8A, comment b; Prosser, Torts (3d Ed.) 31, § 8. In resolving this issue, the court need not ascertain the ultimate reasons why the defendants made certain decisions. For example, in voting to permit black students to remain in Lee and white students to continue at Fifield and O’Hearn, a school committee member might have been prompted by a desire' to improve his prospects for reelection— that would be his motive for so voting, not his intent in so voting. Motive is what prompts a person to take some action toward achieving a result; intent refers to his decision to do a particular act as a means of achieving that result. This is not of course to say that motive may not be relevant to intent; just that they are to be distinguished. Intent ordinarily may not be proved directly, because there is no way of fathoming or scrutinizing the operations of the human mind; but may be inferred from the surrounding circumstances. The opinion of the court in Keyes used the word “motivated” a number of times, at 208-210, but always in the context of the school authorities’ negative burden if plaintiffs should first succeed in proving substantial intentional segregation. To permit school authorities to offer evidence of their own motivations, peculiarly within their own knowledge and control, to rebut plaintiffs’ prima facie case is quite different, in our view, from requiring plaintiffs to adduce evidence of defendants’ motivations.48 A long line of decisions beginning with Brown v. Board of Education, 1954, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873, established that a dual school system explicitly imposed by law is unconstitutional. The Keyes case simply makes it clear that the intentional imposition of a dual school system by covert or subtle means is equally unconstitutional. It should be clearly understood that racial hostility is not the applicable standard. Segregation need not have been inspired by any particular racial attitude to be unconstitutional. Johnson v. San Francisco School District, N.D.Cal.1971, 339 F. Supp. 1315, 1318. Also, the intent to segregate need not be the sole purpose for the defendants’ actions; it need only be one of them.
The question to be decided therefore is whether the defendants have followed a deliberate policy of separating students on the basis of race. Even this statement is an oversimplification since a course of intentionally discriminatory decisions, all arrived at independently, which over the years has produced a dual school system, would be equally unconstitutional although there was no proof of a common scheme linking each decision. Cf. Higgins v. Board of Education, W.D.Mich.1973, No. 6386 (slip op. at 90, 103). This discriminatory purpose may be shown in a *479number of ways. There may be direct evidence of discriminatory purpose, e.g., that segregative steps were taken by the defendants in response to the demands of voters favoring segregation. Similarly, statements by the defendants at the time of decision-making may show discriminatory purpose. Indirect evidence of segregative intent would include the absence of valid educational, fiscal, administrative or other governmental justifications for decisions having clearly foreseeable segregative consequences. Also, in a limited number of situations, e.g., distribution of faculty and staff, an inference of the proscribed intent may arise simply from existing segregation. See Swann v. Charlotte-Mecklenberg Board of Education, 1971, 402 U.S. 1, 18, 91 S.Ct. 1267, 28 L.Ed.2d 554.
The obvious difficulties of ascertaining the intent of the administrators of a municipal school system like Boston’s were mentioned in all four opinions in the Keyes case. Although that case was decided after the trial in these proceedings, the volume and scope of the evidence introduced by the parties and of the requests for findings of fact filed after trial have enabled the court to embark on the teleological course ordered by the Keyes case and charted by Mr. Justice Powell in his opinion, concurring in part and dissenting in part, 413 U.S. at 234-235, as follows:
“Every act of a school board and school administration, and indeed every failure to act where affirmative action is indicated, must now be subject to scrutiny. The most routine decisions with respect to the operation of schools, made almost daily, can affect in varying degrees the extent to which schools are initially segregated, remain in that condition, are desegregated, or — for the long term future— are likely to be one or the other. These decisions include action or non-action with respect to school building construction and location; the timing of building new schools and their size; the closing and consolidation of schools; the drawing or gerrymandering of student attendance zones; the extent to which a neighborhood policy is enforced; the recruitment, promotion and assignment of faculty and supervisory personnel; policies with respect to transfers from one school to another; whether, and to what extent, special schools will be provided, where they will be located, and who will qualify to attend them; the determination of curriculum, including whether there will be ‘tracks’ that lead primarily to college or to vocational training, and the routing of students into these tracks; and even decisions as to social, recreational, and athletic policies.”49
The court’s findings have necessarily been divided into categories, but the ultimate finding as to segregative intent rests upon the record as a whole, including the multiplicity and cumulative effect of the defendants’ policies and practices. A good example of this aspect of the defendants’ conduct appeared in the fall of 1971 after the state board on August 31 approved their amended fourth stage plan which provided for (a) the replacement of open enrollment by controlled transfer, (b) a citizens’ advisory committee to prepare redistrieting plans to achieve racial balance, (c) a comprehensive long-range plan to be formulated with technical assistance to be furnished by the state board and overseen by a special committee, (d) the opening of Lee with a district which would guarantee a balanced school, (e) the opening of Hennigan on a balanced basis, and (f) the construction of a new English which would open balanced. By the end of 1971 the defendants had succeeded in (1) eviscerating the controlled transfer policy by creating exceptions and encouraging their use, (2) sabotaging the advisory committee by appointing persons known to be vehemently opposed to redistricting and instructing them to employ delaying tactics, (3) preventing *480the formulation of a comprehensive long-range plan by refusing the state board access to data necessary to render technical assistance and failing to appoint members to the special committee, (4) permitting a crisis to grow which provided an excuse to open Lee racially segregated and allow children of protesting white parents to remain in Fifield and O’Hearn, and (5) opening one wing of Hennigan on a segregated basis. Also during the 1971-72 school year the defendants were (6) busing 130 black first to third grade students to relatively distant Weld, (7) continuing the existing pattern of overcrowding in predominantly white schools while predominantly black schools had available space, (8) opening new annexes, Charlestown and Carter Temporary, to perpetuate segregation, and (9) rejecting suggestions for the use of portables for the alleviation of racial imbalance.
The record on which the court has relied in determining whether defendants acted with segregative intent also includes several hundred pages of stenographic transcripts of school committee meetings. In some instances excerpts from the transcripts have been included in the court’s findings. The transcripts themselves, however, are far more probative than any excerpts. Unfortunately they present a problem which must be addressed directly if the basis for the court’s conclusions is to be clear, in that they contain numerous degrading characterizations of black students, e.g., “Negro immigrants from the South”,50 by one former, perennial committee member who stated three pages later, “You certainly have a right to say that I seem to be talking about letting the Negroes have their own schools and the whites have their own schools.” The court has placed very little reliance on this type of evidence, which showed beyond doubt the segregative intent of that member, because it did not represent the attitude of most committee members. On the other hand, it has been impossible to overlook it completely, because this member was elected chairman by his colleagues for two years and rarely if ever did any member take exception to his remarks. The court has looked for and weighed valid, nondiscriminatory justifications for the defendants’ decisions and actions. Only when there were none or when there was clear evidence of discriminatory purpose has the court found that the defendants’ intent was segregative. Lastly, the court has endeavored to view the evidence, not with perfect hindsight, but with the perspective and in the chronology and context in which the defendants would have decided upon choices open to them.51
Applying these principles, the court concludes that the defendants took many actions in their official capacities with the purpose and intent to segregate the Boston public schools and that such actions caused current conditions of segregation in the Boston public schools. The findings of fact stated in the preceding divisions of this opinion have described practices which have been ruled to be unconstitutional in many other cases, e. g., Keyes v. School District No. 1, supra, at 201-202; Booker v. Special School District, D.Minn.1972. 351 F.Supp. 799, *481804, 808; Bradley v. Milliken, E.D. Mich.1970, 338 F.Supp. 582, 587-588, 593, aff’d 6 Cir. 1972, 484 F.2d 215; Spangler v. Pasadena Board of Education, C.D.Cal.1970, 311 F.Supp. 501, 508, 520, 522; Taylor v. Board of Education, S.D.N.Y.1961, 191 F.Supp. 181, 184-185, 192, 194, aff’d 2 Cir. 1961, 294 F.2d 36, 38. In five of the categories of defendants’ activities, described in divisions IV, ante, the court concludes on the basis of evidence within each category that the defendants were acting with segregative intent. Only in division VI, ante, on Examination and Vocational Schools and Programs, has the court relied upon the burden-shifting principle discussed in the Keyes case at 413 U.S. 208-210. However, this principle is applicable generally and would be available to buttress the conclusions in divisions I-V if need be. The segregative consequences of many of the defendants’ actions were scarcely contested, but were argued by the defendants to have been unforeseeable or beyond the defendants’ power to prevent. The court has generally rejected these defenses for reasons stated either within numbered divisions of the opinion or in the part entitled Residential Segregation and Neighborhood Schools.
On the issue whether substantial portions of the system have been intentionally segregated by the defendants, the court concludes that they have. Plaintiffs have proved that the defendants intentionally segregated schools at all levels, e.g., secondary English, intermediate Lewenberg and elementary Hennigan; built new schools for a decade with sizes and locations designed to promote segregation ; maintained patterns of overcrowding and underutilization which promoted segregation at 26 schools; and expanded the capacity of approximately 40 schools by means of portables and additions when students could have been assigned to other schools with the effect of reducing racial imbalance. How many students were intentionally separated on a racial basis cannot be stated with any degree of precision; but the annual totals were certainly in the thousands, including graduates of nine K-8 elementary schools and four middle schools by means of feeder patterns manipulated by the defendants, students attending most high schools and several junior highs by the same means, students making imbalancing transfers under the open enrollment policy and exceptions to the controlled transfer policy, students transported to perpetuate segregation, and students at schools identifiably black by means of assignment and transfer policies regarding faculty and staff. As explained in the Keyes case, at 201-202, segregative practices like these have obvious reciprocal effects. For example, by using feeder patterns to channel black students to English, defendants not only concentrated black students there but also made high schools which the black students might otherwise have attended more predominantly white. Similarly every segregative transfer under open enrollment or an exception to the controlled transfer policy, whether by a white or black student, increased segregation in the sending school as well as in the receiving school.
Finally, defendants did not “adduce proof sufficient to support a finding that segregative intent was not among the factors that motivated their actions”, Id., at 210; nor produce “evidence supporting a finding that a lesser degree of segregated schooling” in any part of the system would not have resulted even if they had not acted as they did, Id., at 211; nor demonstrate that' any area of the city “is a separate, identifiable and unrelated section of the school district that should be treated as isolated from the rest of the district”, Id., at 213. In many instances defendants’ evidence consisted of “allegedly logical, racially neutral explanations for their actions”, Id., at 210, which proved to be rationalizations. In view of the plaintiffs’ proof of the defendants’ pervasive practices which were intentionally segregative and their direct and reciprocal effects, the court concludes that the *482defendants’ actions had a segregative impact far beyond the schools which were the immediate subjects of their actions. Indeed plaintiffs’ evidence showed, independently of reciprocal effects, that some of defendants’ practices had a segregative impact on entire levels of the school system. The court concludes that the defendants have knowingly carried out a systematic program of segregation affecting all of the city’s students, teachers and school facilities and have intentionally brought about and maintained a dual school system. Therefore the entire school system of Boston is unconstitutionally segregated. Accordingly, the court will contemporaneously with this opinion file a partial judgment permanently enjoining the city defendants from discriminating upon the basis of race in the operation of the Boston public schools and ordering that they begin forthwith the formulation and implementation of plans to secure for the plaintiffs their constitutional rights.
Remedial Guidelines
It is time to turn to the future. Henceforth the defendants are under an “affirmative obligation” to reverse the consequences of their unconstitutional conduct. Green v. County School Board, 1968, 391 U.S. 430, 88 S. Ct. 1689, 20 L.Ed.2d 716; Raney v. Board of Education, 1968, 391 U.S. 443, 88 S.Ct. 1697, 20 L.Ed.2d 727. Neutral conduct is no longer constitutionally sufficient. Keyes v. School District No. 1, Denver, Colo., supra, 413 U.S. at 200, n. 11. The defendants must eliminate all vestiges of the dual system “root and branch”, Green, supra, 391 U.S. at 438.
In order to assist the defendants in carrying out their obligations, the court will outline several remedial principles which the Supreme Court has declared to be constitutionally applicable. First, the obligation of the defendants is to proceed now to secure the rights of the plaintiffs. Green, supra, at 439. In ascertaining compliance with this obligation, “a school authority’s remedial plan is to be judged by its effectiveness.” Swann v. Charlotte-Mecklenburg Board of Education, 1971, 402 U.S. 1, 25, 91 S.Ct. 1267, 1280, 28 L.Ed.2d 554. Thus, compliance requires desegregation and the goal is to dismantle the dual system. Id. at 28.
Second, the primary responsibility for desegregation lies with the school committee. Brown v. Board of Education, 1955, 349 U.S. 294, 299, 75 S.Ct. 753, 99 L.Ed. 1083 (Brown II). If, in fulfilling this responsibility, policy preferences hinder or obstruct the conversion to a unitary school system, they “must give way [since they would] hinder vindication of federal constitutional guarantees.” North Carolina State Board of Education v. Swann, 1971, 402 U.S. 43, 45, 91 S.Ct. 1284, 28 L.Ed.2d 586. This means that a preference not to bus, or for neighborhood schools, or any other policy preference, can be validly maintained only if it will not interfere with the defendants’ constitutional duty to desegregate. Also “these constitutional principles cannot be allowed to yield simply because of disagreement with them.” Brown II, supra, 349 U.S. at 300. No amount of public or parental opposition will excuse avoidance by school officials of constitutionally imposed obligations.
Third, “school authorities are ‘clearly charged with the affirmative duty to take whatever steps might be necessary,’ ” Swann, supra, 402 U.S. at 15, quoting Green, supra 349 U.S. at 437. This means that busing, the pairing of schools, redistricting with both contiguous and non-contiguous boundary lines, involuntary student and faculty assignments, and all other means, some of which may be distasteful to both school officials and teachers and parents, must be evaluated; and, if necessary to achieve a unitary school system, they must be implemented. The Supreme Court has recognized that “[t]he remedy for such segregation may be administratively awkward, inconvenient, and even bizarre in some situations and may impose burdens on some; but all awkwardness and in*483convenience cannot be avoided . when remedial adjustments are being made to eliminate the dual school systems.” Swann, supra, at 28. The Supreme Court has also indicated that the non-use of a more effective plan may indicate a “lack of good faith” on the part of the defendants, and any such preference places a heavy burden of justification upon them. Green, supra, at 439. With these principles in mind, the difficulty and duration of future proceedings can be kept to a minimum.
Fourth, the Supreme Court has said, “Awareness of the racial composition of the whole school system is likely to be a useful starting point in shaping a remedy to correct past constitutional violations.”' Swann, supra, 402 U.S. at 25. In Boston the public school population is approximately two-thirds white and one-third black; ideally every school in the system would have the same racial proportions, although as a practical matter there is no prospect of-achieving this 2:1 ratio in every school. The Court has also pointed out that with desegregation plans which leave any schools all or predominantly one race, white or black, the defendants must carry the burden that such treatment is genuinely nondiscriminatory. Id. at 26. More specifically, the Court has pointed out that “the nature of the violation determines the scope of the remedy.” Id. at 16. Thus, where the defendants have been found to have coihmitted discriminatory acts as to a specific school facility, certainly no plan which ultimately leaves the racial identifiability of that facility untouched will be constitutionally sufficient.
Fifth, such time as the court allows for formulation and implementation of plans to, desegregate Boston’s schools will not be granted at the cost of continuing the denial of the plaintiffs’ rights under the Constitution. For this reason the time allowed for compliance must be only that reasonably necessary to design and evaluate plans to be presented to the court and thereafter only the time reasonably necessary, administratively, to implement the plan which is ultimately approved by the court. The Supreme Court has decreed that “[t]he burden rests upon the defendants to establish that such time is necessary in the public interest and is consistent with good faith compliance at the earliest practicable date.” Brown II, supra, 349 U.S. at 300.
The remedies to be ordered by the court will be affected in this case by the pending, current implementation by the defendants of a racial imbalance plan designed by the state board and forced upon the city defendants by orders of the Supreme judicial Court entered on January 16, March 22 and April 17, 1974. The state court plan is scheduled to take effect on the opening day of school in September 1974. Our understanding is that this plan will reduce the number of majority black schools from 68 to 44 and the number of black children attending imbalanced schools from approximately 30,000 to approximately 10,000. The state plan relies on two of the traditional methods, redistricting and busing. The plan also removes one of Boston’s_ structural obstacles to integration by converting the entire system, except for McKay junior high in East Boston, to the middle school design, i.e., all students entering the sixth grade will be in a middle school and all students entering the ninth grade will be in a high school except at McKay. Under the state board plah, Boston’s high schools, except for the three examination schools, are given districts for the first time, thus eliminating many discriminatory feeder patterns. Finally, busing will be used to reduce segregation; state officials predict that approximately 6,000 students will be bused for this reason, but city officials estimate that the number will be closer to 20,000. These and other aspects of the state plan will be explored at the hearing scheduled by. separate notice for June 27, 1974 in the context of the court’s separate partial judgment and interlocutory order, attached hereto as Appendix A.
*484APPENDIX A
PARTIAL JUDGMENT AND INTERLOCUTORY ORDER
This action came on for trial before the court, and the issues having been duly tried and a decision having been duly rendered, as set forth in an opinion filed herewith, that the rights of the plaintiff class of black students and parents under the Fourteenth Amendment to the Constitution of the United States have been and are being violated by the defendants in their management and operation of the public schools of the City of Boston,
Partial Judgment
It is Ordered and Adjudged that the defendants John J. Kerrigan, chairman, Paul Ellison, John J. McDonough, Kathleen Sullivan and Paul R; Tierney, who are the members of the Boston School Committee, and William J. Leary, Superintendent of Schools of the City of Boston, their officers, agents, servants, employees, attorneys, and all other persons in active concert or participation with them who receive actual notice of this judgment and order, be permanently enjoined from discriminating upon the basis of race in the operation of the public schools of the City of Boston and from creating, promoting or maintaining racial segregation in any school or other facility in the Boston school system; and said defendants are further Ordered to begin forthwith the formulation and implementation of plans which shall eliminate every form of racial segregation in the public schools of Boston, including all consequences and vestiges of segregation previously practiced by the defendants.
Interlocutory Order
It is further Ordered that, pending further order or unless specific leave of this court is obtained, said defendants be preliminarily enjoined from:
(a) failing to comply in any respect with the Racial Imbalance. Act plan ordered by the Supreme Judicial Court of Massachusetts to be implemented on or before the opening day of school in September, 1974;
(b) beginning the construction of any new school or expansion or the placement of any new portable;
(c) granting transfers of white teachers from schools with majority black enrollments or black teachers from schools^ with majority white enrollments; t
(d) granting transfers under exceptions to the controlled transfer policy.