OPINION

PECKHAM, District Judge.

I. PROCEDURAL HISTORY

Plaintiffs, parents of Spanish-surnamed American children attending public school in the San Jose Unified School District, bring this action on behalf of themselves and all others similarly situated. They charge that defendants, the San Jose Unified School District, its Superintendent, and members of its Board of Education, have purposefully operated and maintained a segregated public school system in violation of the Fourteenth Amendment. Defendants admit the existing racial imbalance in the school system, but argue that ethnic imbalance results not from defendants’ intentional conduct but from demographic and residential patterns over which they have no control. Jurisdiction is vested in this court by 28 U.S.C. §§ 1343(3), 1343(4) and 42 U.S.C. § 1983.

On November 4, 1971, plaintiffs instituted this action and sought a temporary restraining order and a preliminary injunction to restrain defendants from completing construction of the following schools until the court could determine the merits of this case:

College Park Elementary

Anne Darling Elementary

Gardner Elementary

Hester Elementary

Jefferson Elementary

Lincoln Glen Elementary 1

Longfellow-Edison

(combined elementary/continuation) 2

Lowell Elementary

Mann Elementary

Olinder Elementary

Washington Elementary

Roosevelt Junior High

Wilson Junior High

On November 4, this court denied plaintiffs’ ex parte application for a temporary restraining order and issued an order to show cause. After a hearing on November 10, 1971, the court issued a temporary restraining order prohibiting construction of almost all 3 of the schools slated for reconstruction pending a hearing on plaintiffs’ motion for preliminary injunction.

After extensive oral argument on December 17, 1971, this court denied plaintiffs’ motion for a preliminary injunction. In the opinion dated January 12, 1972, the court concluded that on the record presently available:

. [t]his Court does not believe that such [ethnic] imbalance alone constitutes a violation of plaintiffs’ civil rights. As the Sixth Circuit recently pointed out in Davis v. School District of City of Pontiac, Inc., 443 F.2d 573 (6th Cir.), cert. denied, 40 U.S.L.W. 3192 [404 U.S. 913, 92 S.Ct. 233, 30 L.Ed.2d 186] (1971), ‘a school district has no affirmative obligation to achieve a balance of the races in the *312 schools when the existing imbalance is not attributable to school policies or practices and is the result of housing patterns and other forces over which the school administration had no control.’ 443 F.2d at 575.

Plaintiffs cannot hope to succeed on the merits, then, unless they can demonstrate how the defendant School District has acted to create ethnic imbalance in the San Jose schools. After examining the evidence plaintiffs have offered in support of a preliminary injunction this ' Court can find no instances where defendant School District has acted impermissibly. The evidence shows that defendant School District has adhered to a ‘neighborhood school policy,’ with the result that ethnic composition of the schools merely reflects residential patterns. This is not a case like Spangler v. Pasadena City Bd. of Education, 311 F.Supp. 501 (D.Colo.1970); Keyes v. School District Number 1, Denver, Colo., 313 F.Supp. 61 (D.Colo.1970), aff’d in part, rev’d in part, 445 F.2d 990 (10th Cir. 1971); or Davis v. School District of City of Pontiac, Inc., supra, where the defendant school districts claimed to be following a neighborhood school policy, but conveniently refrained from strict adherence to that policy when it appeared to be leading to increased integration.

Accordingly, the Court finds that given the evidence available at this stage in the litigation, there is an insufficient likelihood that plaintiffs will prevail on the merits to warrant the issuance of a preliminary injunction; and plaintiffs’ motion is hereby DENIED.

Trial on the merits was held on July 16, 18 — 19, 23 — 25 and August 6, 8 — 9, 16 and 20, 1974. A transcript was prepared and the parties were ordered to submit proposed findings of fact and conclusions of law. The court heard oral argument on April 30, May 1 and May 13, 1975.

II. THE PARTIES

Plaintiffs in this action are Arnulfo M. Diaz and Socorro Diaz, as parents and next friends of Fernando Diaz, Miguel A. Diaz and Juan F. Diaz; Jose Vasquez, as parent and next friend of David Vasquez, Susan Vasquez and Jody-Lynn Vasquez.

When this action was filed in November, 1971, each of the six named minor plaintiffs attended either junior high school or high school within defendant San Jose Unified School District. By 1973, when the trial commenced, two of the student plaintiffs attended either junior high school, high school, or continuation schools, and both were expected to enroll and continue in San Jose high schools for the school year 1974r-1975.

Defendants are the San Jose Unified School District (“school district”), a political subdivision and school district of the State of California; Charles Knight, individually and as District Superintendent of the San Jose Unified School District; Neil H. Geier, Jr.; Elizabeth J. Allen; Edwin P. Jones, Jr.; Mary K. McCreath; and Donald L. Raimondi, individually and as members of the Board of Education of the San Jose Unified School District (“board”) as of November, 1971, when this action commenced. During the pendency of this action, Phillip L. Hammer replaced Edwin P. Jones, Jr. as a board member, effective July 1, 1973, and having appeared as a party herein through counsel of record is hereby substituted as a party in place of Mr. Jones.

This is a class action brought pursuant to Rule 23, Federal Rules of Civil Procedure, the class consisting of all Spanishsurnamed students enrolled in the San Jose Unified School District and their parents. Spanish-surnamed Americans are the only substantial minority group in the district.

III. GENERAL BACKGROUND

San Jose Unified School District is a political subdivision of the State of California operated by and pursuant to state law. General direction and control of the district is vested in the Board of

*313 Education comprised of five members who are elected to four-year terms. The Chief Executive Officer of the school district is the Superintendent of Schools.

From 1963 through 1974, the following persons held positions as Superintendent and board members of the district:

Superintendent Board Members

1963 Crandall Jones Gunderson Gale Butcher Douglass

1964 Crandall Jones Gunderson Gale Butcher Geier

(8/6/64)

1965 Crandall Jones Gunderson Gale Butcher Geier

1966 Downing Jones Gunderson McCreath Butcher Geier

(7/1/66)

1967 Downing Jones Gunderson McCreath Butcher Geier

1968 Downing Jones Gunderson McCreath Butcher Geier

1969 Downing Jones Hopkins McCreath Raimondi Geier

(5/1/69) (7/1/69)

1970 Knight Jones Hopkins McCreath Raimondi Geier

(7/1/70)

1971 Knight Jones Allen McCreath Raimondi Geier

(7/1/71)

1972 Knight Jones Allen McCreath Raimondi Geier

1973 Knight Hammer Allen McCreath Raimondi Geier

(7/1/73)

1974 Knight Hammer Allen McCreath Raimondi Geier

The district is situated in Santa Clara County, with the northerly terminus being located approximately at the intersections of Highway 17 and Highway 101 (Bayshore Freeway) in the City of San Jose, California. It extends south in a banana-like shape through central San Jose, approximately 16 4 air miles to a point beyond the Almadén Valley area. The width of the district varies along its length from four air miles maximum to one and one-half air miles minimum across in an east-west direction, the narrowest portion being located approximately at the southern boundaries of Broadway and Washington Elementary Schools. The driving distance of the district totals almost 20 miles. The ex-tremé north and south schools, Bachrodt and Hacienda, while only 12.2 air miles apart, are separated by 17 miles’ driving distance along the major north-south transportation route which winds through central San Jose.

The northern one-half of the district is composed of residential neighborhoods, the San Jose Municipal Airport, a Civic Center and central San Jose. The makeup of the area has been altered by urban renewal, freeway construction and acquisition, extension of industrial, commercial and office facilities, and expansion of the airport. The southern half of the district is primarily a suburban area which developed from the late 1950’s throughout the 1960’s.

Since World War II, ethnic minorities have been increasingly concentrated in San Jose’s northern downtown area. While the total population in the downtown area increased only slightly between 1950 and 1960, the minority population more than tripled.

*314 In 1950, the school district consisted of a total of 24 schools: 16 elementary schools, 5 junior high schools, and 3 high schools. The northern boundaries of the district were extended by the 1956 annexation of the Sunol School District. The southern boundaries were extended by the 1956 annexation of the Almadén Elementary School District. Since 1956, when Sunol and Almadén were annexed, the external boundaries of the district have not changed.

Between 1950 and 1970, the student enrollment in the district increased by over 250%, and the total number of schools doubled to house the increased student population. By the fall of 1973, the district’s Board of Trustees operated 43 schools, enrolling 36,687 students. There are presently 31 elementary schools, three 3-year junior high schools, two 2-year junior high schools, three 3-year high schools, two 4-year high schools, and two special schools.

The racial composition of the district student population is as follows:

TABLE 1

Race Percentage Total

Anglo 70.8 25,974

Spanish-surnamed 24.6 9,037

Black 1.5 561

Other Minorities 3.0 1,118

A “segregated” or “ethnically imbalanced” school is defined as a school in which the racial-or ethnic makeup of the school’s student enrollment is substantially out of proportion with the racial and ethnic makeup of the student population as a whole. Under guidelines promulgated by the State of California, a school is imbalanced “if the percentage of pupils of one or more ethnic groups differs by more than 15 percentage points from that in all schools in the district.” Calif.Admin.Code § 2011.

Since 1964, the schools of the San Jose Unified School District have been substantially imbalanced. The State Department of Education Bureau of Intergroup Relations advised the district in the fall of 1968 that 41 of its 50 schools were ethnically imbalanced. The trend toward concentration has continued: while the total student enrollment in the northern downtown schools has declined, the percentage of Spanish-surnamed students has increased.

The Spanish-surnamed population in the district has declined from 29.7% in 1966 to 24.6% in 1973. From this group, 78.8% of the Spanish-surnamed school-age children reside in the northern downtown area. Reflecting this demographic concentration, the student population of the northern downtown schools is presently 60.5% Spanish-surnamed, while the student population of the southern schools is only .07% Spanishsurnamed.

The district has followed a pattern of selecting sites for school construction according to their projected long-range requirements. For additional capacity to meet peak year enrollments, portable classrooms are attached to permanent structures or tacked together to form independent units.

Student assignment is dictated by a mandatory attendance boundary; the board requires students to attend schools within the designated geographical areas of their residence. This requirement, commonly termed a “neighborhood school policy,” fixes boundaries for schools so that they are relatively convenient to the students; the goal of the system is “walking distance” for as many children as enrollment and facility availability permit. Because “walking distance” varies substantially between elementary and secondary pupils, the board uses transportation within attendance areas where mandated by considerations of distance and safety. 5 The only students transported outside attendance areas are those with special needs — e.g., those with physical or mental disabili *315 ty — who attend special schools which serve the entire district.

IV. FACTS

To prove that defendants have operated a segregated school system in violation of the Fourteenth Amendment, plaintiffs first established that the board was fully aware of the existing racial imbalance in the district, and then attempted to demonstrate that ethnic imbalance resulted from defendants’ intentional conduct. To prove segregative intent, plaintiffs introduced evidence of the district’s educational policies on (1) site selection and school construction; (2) adoption of a neighborhood school policy with board-designated attendance areas; (3) reconstruction of Field Act schools; (4) school closures and student reassignments; (5) location of portable classrooms and maintenance of double sessions; (6) student transportation; (7) presentation of materials supporting bond elections; (8) response to integration proposals by a citizens’ committee; (9) faculty and staff assignments and (10) failure to integrate despite a state statutory duty and a publicly-issued board policy to relieve ethnic imbalance.

A. District’s Knowledge of Racial Imbalance

The district’s board of education has long been aware of the growing concentration of minority residents in the northern downtown area. The board acknowledged the existence of segregated housing patterns in a 1963 resolution. Moreover, the San Jose Planning Department prepared a public report entitled “Housing Report for the Planning Commission,” detailing the ethnic residential makeup of the district and projecting racial demographic patterns through 1990. In 1970, the San Jose City Planning Department prepared a “Demographic Report” for a subcommittee of the district’s Quality Urban Education Study Team (QUEST), a citizens’ committee charged with evaluating integration proposals in the San Jose school system. The board had access to these reports and knew their recommendations and supporting data.

The board has been and is aware that it operates a system of predominantly Spanish-sumamed schools in the northern downtown core and predominantly Anglo schools in the southern suburban area.

As early as 1962, the board acknowledged that segregation was inherently harmful and recognized the need for a comprehensive study of racial and ethnic isolation within the schools. In a resolution dated January 17, 1963, the board recognized the correlation between de facto segregation and social evils affecting community life. Stating that segregation is aggravated by designation of school boundaries, the board promised to initiate a comprehensive study of ethnic segregation in the school system.

In 1964, the district conducted its first ethnic survey of the schools which confirmed the pattern of imbalance in the schools. In 1965, in what was then policy # 5007.1, the board pledged that, in establishing attendance areas, it would attempt to avoid ethnic segregation of students.

Starting in 1966, in compliance with state requirements, the district conducted annual racial and ethnic surveys of its schools, which disclosed in each and every year a continued pattern of racial imbalance.

In 1968, the Santa Clara County Office of Education issued a report entitled “Improving Ethnic Balance and Inter-group Relations” prepared by consultant Edwin T. Rios, which discussed the deleterious effect of ethnic imbalance in the public school system.

The State Department of Education advised the board that in 1968 41 of its 50 schools were racially imbalanced by state standards. By 1973, substantial imbalance remained in the public schools.

On February 10, 1969, the board unanimously voted to reaffirm its commitment to a quality integrated educational program. The board also voted to establish a broad-based citizens’ committee *316 charged with launching a comprehensive study of ethnic imbalance in the school district and evaluating proposals for integrated education. The board authorized the newly-formed committee to work with consulting services from the State Department of Education and other agencies. By this resolution, the board officially created the QUEST committee.

On March 11, 1970, the board issued a policy statement directed to the QUEST committee, reaffirming its commitment to integration:

To us a commitment to quality integration means equal opportunity for each student in the district to obtain a quality education. To many, the emphasis has been only on “quality education.”

In this regard, we are mindful of a recent statement of President Nixon. “ ‘Quality education’ is already being interpreted as a code word for a delay of desegregation. We must never let this meaning take hold.”

Quality is what education is all about; desegregation is vital to that quality; as we improve the quality of education for all American children, we will help them improve the quality of their own lives in the next generation.

We, too, believe that integration of the schools is a necessary and important component of quality education. Therefore, it is our desire that Quest review all attendance practices including busing, maintenance of neighborhood schools, transfer policies, attendance boundaries, etc., in light of how they add or detract in the accomplishment of our primary objective of a quality integrated education.

However, integration as the “primary objective” is modified later in this same policy statement.

On the problem as a whole, however, it is our preliminary judgment that as between the broad areas of instructional program on the one hand and achieving racial and ethnic balance on the other for the near term priority should be given to the instructional program. We are motivated in this by the belief that the hoped-for benefits of racial and ethnic balance in the schools will not be achieved unless revisions have been made in the instructional program, the grouping of students and the training of teachers to make the entire school experience more responsive to the needs of all elements of our community. Simply stated, a racially balanced school is not necessarily an integrated one and does not necessarily provide a quality education. For the long term, we feel that priority should be placed on achieving racial and ethnic balance in our schools to the fullest extent possible within our means as it is to us an essential component of quality integrated education.

Finally, we wish to restate our overall philosophy that we are committed to serving the needs of this community. And as in the past, this Board would not voluntarily adopt a program which it had reason to believe the community had not participated in or would not support.

As indicated by the events listed above, the board clearly knew of the established pattern of racial imbalance within the school system for years prior to the filing of this action. Defendants readily concede both the existence of ethnic imbalance and their awareness of the problem. The central issue in this litigation is whether the board intentionally created or maintained ethnic imbalance.

B. District Actions and Policies

1. Site selection and school construction. Since 1965, the district has constructed nine new schools on newly-selected sites. As shown by Table 2, all nine opened as imbalanced schools. In selecting sites for school construction in the Spanish-surnamed northern downtown core, the school district allegedly followed the guidelines outlined in the state’s “Checklist for Evaluation of Attendance Practices and/or Proposed School Sites”; however, the district *317 failed to consider the state guidelines for school construction in the Anglo suburban area. When it selected sites, the board knew that the newly-constructed schools would be imbalanced from inception.

By 1973, the district had planned or commenced construction of additional schools in the predominantly Anglo southern suburban area. Ethnic residential patterns indicate that these facilities will open as imbalanced schools.

TABLE 2

Year School Opened Percentage Percentage Spanish- Spanlshsurnamed surnamed When Opened in 1973

Williams (elementary) 1968 2.2 1.4

Henderson (elementary) 1966 5.4 1.5

Slmonds (elementary) 1966 1.7 2.7

Bret Harte (Jr. high) 1966 7.0 3.4

Leland (high) 1967 5.1 3.4

Allen (elementary) 1965 14.6 • 4.7

Erickson (elementary) 1968 9.4 6.4

Hammer (elementary) 1966 13.7 6.6

Terrell (elementary) 1966 20.8 7.7

2. Replacement of Field Act schools. The Field Act was adopted by the California Legislature in 1933 (former California Education Code § 18191), and is presently found in California Education Code §§ 15451 et seq. and § 15501 et seq. This “emergency” legislation resulted from public concern following the destruction of many schools by the Long Beach earthquake on March 10, 1933. The rules and regulations prescribed un-' der the authority of the Field Act established minimum requirements for the design, construction and reconstruction of public school buildings “in order to attain the requisite stability to withstand vertical loads and lateral forces (wind or earthquake) . . . ” 20 Cal. Admin. Code, § 2.

Under the 1967 Amendments to California Education Code, §§ 15501 et seq., all school districts were required to order an inspection of school facilities on or before January 1, 1970, to evaluate the safety of buildings. The amended statute required a district governing board to commence plans for the repair, reconstruction, or replacement of any building found unsafe no later than six months from receipt of the examination report.

After the statutory inspection, structural engineers found that numerous school buildings in the San Jose school district failed to meet the requirements of the Field Act. 6

Following the Los Angeles earthquake in February, 1971, which, among other casualties, destroyed several school buildings, the board voted to close all unsafe Field Act schools at the end of the 1971 school year. The district ordered the buildings razed, and, pending reconstruction, housed the students in portable structures with double sessions.

On January 21, 1971, the board created a Building Master Plan Task Group (“Task Group”) and subsidiary local committees with responsibility for individual high school attendance areas (“Task Forces”). The Task Group’s goal was to establish a construction program and launch a “fast-track” building program to house the Field Act students in new facilities by fall of 1972. In formulating recommendations for the district’s reconstruction policy, the Task Group studied the proposed reconstruction of Field Act schools on existing sites. The Task Group commenced work on April 14, 1971, and continued with assistance from the Westinghouse Learning Corporation.

Shortly after receipt of the Task Group’s report, in May of 1972, the board decided to rebuild most of the downtown Field Act schools on their original sites. *318 The decision reflected the demand of local parents; the board did not consider the impact of reconstruction on existing ethnic imbalance. No legal obligation compelled the board to rebuild the Field Act schools on their original sites. On August 5, 1971, the board resolved to authorize construction of the following facilities in replacement of Field Act schools:

(a) Schools were to be built in the Washington-Gardner area;

(b) Olinder School was to be rebuilt on the existing Roosevelt site;

The board later authorized the following changes or additions to the basic plan:

(1) Olinder was rebuilt on a “new site”;

(2) Mann remained rebuilt as a portable on site; and

(3) Lincoln Glen (new classroom space) was rebuilt.

Local committees of the Task Group were subjected to opposition to reconstruction of the Washington and Gardner Elementary schools and Wilson Junior High School; however, members of the Spanish-surnamed community pressured for rebuilding of these facilities. Subsequently, over objection of board member McCreath — who voiced a concern that rebuilding of Wilson Junior High might constitute an act of de jure segregation — the board approved reconstruction of Gardner and Washington elementary schools and of Wilson Junior High School, with a restructured grade level. (Wilson was ultimately not rebuilt. See discussion, infra.)

The board’s election to rebuild most Field Act schools on their original sites resulted in the following pattern of ethnic imbalance in the student population:

TABLE 3

Rebuilt Field Act Schools 1971 Percentage SpanlshSchool surnamed 1973 Percentage Spanishsurnamed

Washington 78.4 86.6

Gardner 87.8 Grant 71.7 86.5 78.9

Olinder (traded with city

for land) 73.8 76.2

Anne Darling 73.4 67.4

Burnett 65.2 66.6

Hester 43.6 59.2

Lowell 49.5 58.4

Horace Mann 50.2 56.0

Trace 18.3 24.0

Willow Glen 16.8 16.3

Almadén (site moved from commercial area) 6.3 6.5

Trace and Willow Glen were the only two schools that were balanced in 1971 and the only two that were balanced in 1973. Ten of the reconstructed schools were imbalanced in 1971 and remained imbalanced in 1973.

Only two of the Field Act schools, Olinder and Almadén, were rebuilt on different sites. The relocation failed to relieve imbalance; both schools were imbalanced in 1971 and remained so in 1973. In neither school were attendance areas or boundaries altered.

In 1971, the student population at Washington Elementary School was 78.4% Spanish-surnamed and the student population at Gardner was 87.8% Spanish-surnamed. The two schools, with the district’s greatest Spanish-surnamed concentration in 1971, remained the district’s most imbalanced schools in 1973. When the board directed the schools closed under the Field Act, the board established portable classrooms at the original sites and maintained the existing boundaries. Recognizing that Washington and Gardner would remain imbalanced schools, the board voted over member McCreath’s opposition to rebuild the facilities on their original sites:

Q. What was your opposition, Mrs. McCreath?

*319 A. The Washington-Gardner-Wilson area is a pocket area, it’s pretty well self-contained, it’s cut off by a freeway and it has traditionally been an area that is very much in imbalance, about as imbalanced as any in our district.

Q. By that you mean segregated?

A. Yes, segregated. And I felt that that is one area in our school system that is close to Willow Glen, the transportation would not be difficult, and I felt that we should begin integrating in that area.

Q. In other words, the students could attend Willow Glen easily?

A. Yes. However, community pressure was brought to bear to maintain schools, replace the schools.

Q. On their original sites?

A. On their original sites.

By 1973, Washington — constructed entirely from portable classrooms — serviced 797 students, of whom' 86.6% were Spanish-surnamed. Because of the crowding, Washington students attended double sessions.

In electing to rebuild Washington and Gardner, the board rejected a Westinghouse Study proposal recommending that Washington and Gardner students be dispersed to contiguous schools. Dr. Knight testified that adoption of the Westinghouse Study proposal would have increased ethnic balance among the area’s schools. There existed several alternative methods for student reassignment which would have reduced imbalance. One mile from Washington, the Riverglen School housed 382 students on a 7.6 acre site. With an enrollment of 19.2% Spanish-surnamed students, River-glen was 67.4% less concentrated than Washington. Despite the fact that Riverglen was operating under capacity, no Washington students were transferred. Students from the Little Orchard neighborhood, which is almost 100% Spanishsurnamed, were assigned for years to Washington, even though they actually lived closer to Riverglen. Although the board considered plans to change boundaries for several years, it maintained the existing lines in the face of opposition from Riverglen parents.

Canoas, situated approximately three or four miles from Gardner, was housing 218 students — only 9.7% of whom were Spanish-surnamed — on a 7.25 acre site. With no portables, Canoas was operating under capacity. At no time did the board consider transferring Gardner students to Canoas to alleviate imbalance. Lincoln Glen, located on a 6.4 acre site one mile from Washington and two miles from Gardner, housed 326 students, of whom 15% were Spanish-surnamed. Although Lincoln Glen was operating under capacity, the board never considered transferring students to Lincoln Glen to relieve imbalance at Washington or Gardner.

3. School closures. Four of the Field Act schools were not rebuilt: Jefferson, Longfellow, Wilson and Roosevelt. Three downtown non-Field Act schools were also closed: College Park, Belden and Bascom. Reassignment of students from the closed schools generally resulted in newly-constituted student enrollments with better ethnic balance than the existing racial concentration at the original schools. In each case, the Spanish-surnamed population increased at the receiving school.

Effective September, 1972, the district consolidated the Longfellow and Hester attendance areas, thereby eliminating Longfellow, which had a 90.4% Spanishsurnamed enrollment in 1971; the consolidated student body resulted in a 60.7% Spanish-surnamed concentration at Hester. A school-by-school comparison of possible combinations of Longfellow with adjacent elementary attendance boundaries (using 1971 figures) indicates that combination with Hester resulted in the best ethnic balance at the receiving school:

*320 Spanish- Spanishsurnamed surnamed Enrollment Enrollment Percentage

Longfellow: 136 123 90.4

1. Hester 419 555 186 44,3 309 55.7 (best)

2. Bascom 174 310 73 39.9 196 63.2

3. Jefferson 339 475 247 72.9 370 77.9

4. Mann 451 587 240 53,2 363 61.8

5. Washington 828 964 644 77.8 767 79.6

Effective September, 1972, the district combined the attendance areas of Bascom and Trace. 7 The Spanish-surnamed student enrollment, which had totalled 39.9% at Bascom, dropped to 24% in 1972 at Trace. A comparison of the possible combinations of Bascom with adjacent elementary district boundaries indicates that the combination with Trace resulted in the best ethnic balance at the receiving school:

Spanish- Spanishsurnamed surnamed Enrollment Enrollment Percentage

Bascom: 174 73 39.9

1. Trace 434 608 81 18,7 154 25.3 (best)

2. Hester 419 593 186 44.3 259 43.6

3. Gardner 522 701 462 87,6 535 76.3

4. Broadway 505 679 204 40.4 277 40.8

5. Longfellow 136 310 123 90.4 196 63.2

By resolutions dated February 15, 1973, and April 26, 1973, effective Fall of 1973, the board consolidated the attendance areas of College Park, Belden, Jefferson, and Bachrodt, creating a new Bachrodt attendance area. 8 In electing to assign the primarily Spanish-surnamed students to Bachrodt, the board failed to consider bussing students to southern suburban schools and failed to formulate alternative proposals directed toward relieving ethnic imbalance. The board’s inquiry was limited to redesignating boundaries within the northern downtown area; boundary alteration within this area would not affect racial concentration.

Art Ryder, a parent and former chairperson of the Magnet School Task Force of QUEST, stated in an appearance before the board on March 15, 1973, that closing College Park and Jefferson afforded an opportunity to implement integration. He urged the board to transfer College Park and Jefferson students not to Bachrodt but to schools south of Willow Glen.

The board, however, failed to consider the Ryder proposals or investigate alternative measures directed towards improving ethnic balance. The board decided to keep the students “in generally the same area,” rather than relieve imbalance.

A school-by-school analysis of the closed facilities (1972 figures) indicates that consolidation of each school with Bachrodt resulted in a better ethnic balance than possible by combination with any adjacent attendance areas.

Enrollment. Spanish- Spanishsurnamed surnamed Enrollment Percentage

College Park: 221 136 61.5

1. Hester 526 747 311 59.1 447 59.8

2. Jefferson 285 506 203 71.2 339 67.0

3. Bachrodt 225 446 67 29,8 203 45.5 (best)

4. Belden 291 512 175 60.1 311 60.7

Jefferson: 285 203 71.2

1. Hester 526 811 311 59.1 514 63.4

2. College Park 221 506 136 61,5 339 67.0

3. Bachrodt 225 510 67 29.8 270 52.9 (best)

4. Belden 291 576 175 60.1 378 65.6

5. Grant 644 929 486 75.5 689 74.2

*321 Spanish- Spanishsurnamed surnamed Jefferson (C't'd) Enrollment Enrollment Percentage

53.1 6. Mann 469 754 249 452 59.9

Belden: 291 175 60.1

1. Bachrodt 225 516 67 242 29.8 46.9 (best)

2. Grant 644 935 486 661 75.5 70.7

3. Jefferson 285 576 203 378 71.2 65.6

Based on the 1972 student ethnic ratios, the projected Spanish-surnamed percentage of Bachrodt students in 1973 would have increased from 29.8% to 56.6%. This would have represented a 5% decrease for College Park students, a 4% decrease for Belden students and a 15% decrease for Jefferson students; no superior alternative combination was possible within the adjacent attendance areas. The actual (as opposed to projected) percentage of Spanish-surnamed students increased to 61.9%; this resulted from a departure of approximately 180 students —150 of whom were Anglo — from the combined attendance areas.

In 1971, the board closed Roosevelt Junior High School, with 64.8% Spanishsurnamed students, and Wilson Junior High School, with 79.6% Spanish-surnamed students. Both had been three-year schools servicing seventh, eighth and ninth grade students.

After the disputed decision of October 21, 1971, to reconstruct Wilson, the district unsuccessfully attempted to obtain facilities for Wilson in the September 1972 bond election. On December 21, 1972, the Woodrow Wilson Rebuilding Committee reminded the board of its obligation to rebuild Wilson; however, in the 1973 bond issue, the district failed to include funds for a reconstructed Wilson. Following a series of actions initiated by the Superintendent on July 3, 1973, and a series of board discussions on July 12, July 19, September 16 and September 20, the board decided not to rebuild Wilson. Instead, the board finalized the existing interim allocation of junior high school students among Markham, Burnett and Hoover junior high schools and Lincoln High School.

The two remaining downtown junior high schools, Burnett with 64.9% Spanish-surnamed students and Hoover with 27.8% Spanish-surnamed students, were contracted from three-year (seventh, eighth, and ninth grades) junior high schools to two-year schools (seventh and eighth grade). Correspondingly, two downtown high schools, Lincoln High School, with 43.1% Spanish-surnamed students, and San Jose High School, with 59.6% Spanish-surnamed students, received the ninth grades formerly housed at Burnett and Hoover and became four-year high schools servicing ninth through twelfth grades.

This restructuring resulted in the following attendance pattern. Seventh and eighth grade students, formerly attending Wilson Junior High School, now attended Hoover, Burnett or Markham junior high schools; former Wilson ninth grade students now attended Lincoln High School. Seventh and eighth grade students formerly attending Roosevelt Junior High School now attended Burnett Junior High School; former Roosevelt ninth grade students now attended San Jose High School.

The restructuring resulted in the following ethnic impact on the receiving schools:

TABLE 4

Percentage Percentage Spanish- Spanishsurnamed surnamed School (1970) (1973)

Burnett Junior High 64.9 66.9

Hoover Junior High 27.8 42.8

San Jose High 59.6 64.7

Lincoln High 43.1 50.2

Markham Junior High 7.2 18.4

Thus, restructuring aggravated racial imbalance at each receiving school. The most pronounced case was Hoover Junior High School which, formerly one of the district’s few ethnically balanced schools, became an imbalanced school. From the vantage point of the transferred stu *322 dents, however, restructured attendance either improved ethnic balance or maintained the previous level. Seventh and eighth grade students at Roosevelt transferred from a school with a 64.8% imbalance to a school with a 66.9% imbalance. Roosevelt ninth grade students transferred from an imbalance of 64.8% to an imbalance of 64.7% at San Jose High School. Seventh and eighth grade students at Wilson Junior High School transferred from a school with a 79.6% imbalance to schools with imbalances of 42.8% (Hoover Junior High School), 66.9% (Burnett Junior High School), and 18.4% (Markham Junior High School). Ninth grade students at Wilson transferred from 79.6% imbalance to 50.2% imbalance at Lincoln High School.

The board failed to formulate or évaluate alternative proposals for student transfers from Roosevelt. The board did not consider transporting Roosevelt students to a less impacted facility than San Jose High School. Nor did the board consider housing Roosevelt Junior School students in portables at Bret Harte Junior High School or John Muir Junior High School in the south, which would have resulted in improved balance. In housing high school students transferred from Roosevelt, the district did add portables to San Jose High School in the northern downtown core.

Before finalizing the restructuring decision, the district considered transferring all Wilson Junior High School students to adjacent Markham Junior High School, with only 7.2% Spanish-surnamed students. Because Markham was conducting double sessions, the proposed transfer was contemplated as a temporary measure pending location of a site in the southernmost portion of the northern downtown core for a magnet school with a large “integrated student body.” However, community opposition developed, and a parents’ meeting held at Markham Junior High School indicated mass hostility to bussing students from Wilson to Markham. In the face of this opposition, the board decided to restructure grades and transfer the majority of Wilson students to the northern Hoover, Burnett and Lincoln junior high schools.

By 1973, the junior and senior high schools in the district reflected the following ethnic makeup:

TABLE 5

Percentage School Spanish-surnamed

Four-Year High Schools

San Jose (serves northeast) 64.7

Lincoln (serves northwest) 50.2

Three-Year High Schools

Willow Glen (serves attendance area crossing entire district, south of San Jose and Lincoln attendance areas) 13.6

Pioneer (serves attendance area crossing entire district, south of Willow Glen) 8.1

Leland (serves south) 3.4

Two-Year Junior High Schools

Burnett (serves northeast) 66.9

Hoover (serves northwest) 42.8

Three-Year Junior High Schools

Markham (serves attendance area crossing entire district, south of Burnett and Hoover) 18.4

Muir (serves attendance area crossing entire district, south of Markham) 8.6

Bret Harte (serves south) 3.4

4. Assignment of portables and administration of double sessions. Portable classrooms are moveable and designed for quick temporary service. The district owns over 400 portable classrooms which are distributed throughout 39 of the district’s 42 schools. In addition, the district leased 112 portable classrooms to house students pending reconstruction of Field Act schools.

In assigning portable classrooms, the district has “followed a policy of picking them up and moving them around to schools to accommodate whatever the needs are at that particular school.” The district has never used portable location to improve ethnic balance.

The district has used portables to create schools on sites in the imbalanced northern downtown area. In the Fall of 1973, seven downtown schools were totally constructed from portable class *323 rooms. No consideration has been given to transporting northern students to a southern suburban school.

Students in the southern suburban part of the district have been overcrowded in their present facilities. As a temporary measure, until new construction is completed, the southern schools have conducted double sessions. The double-session relief measure did not alter the ethnic makeup of the southern schools, which, with a heavy concentration of Anglo students, are ethnically imbalanced.

In the school year 1973 — 1974, the two junior high schools located at the southernmost part of the district conducted double sessions. John Muir High School, with a student capacity of 1,554, serviced 2,167 students; Bret Harte Junior High School, with a student capacity of 1,764, serviced 2,220 students. Both southern junior high schools were severely imbalanced: the percentage of Spanish-surnamed students totalled 8.6% at John Muir and 3.4% at Bret Harte. The two northern junior high schools remaining after closure of Roosevelt and Wilson were also imbalanced: the student enrollment at Burnett was 66.9% Spanishsurnamed, while the student population at Hoover was 42.8% Spanish-surnamed. Thus, the district was operating four ethnically imbalanced junior high schools. The fifth, Markham Junior High School, with 18.4% Spanish-surnamed students, approached racial balance. The district could have relieved congestion in the southern schools and improved ethnic balance in all junior high schools by transporting students from southern facilities to the undercapacity northern schools.

At a school board meeting on May 2, 1974, a group of Anglo parents demanded that their children be relieved of double sessions at Muir and Bret Harte junior high schools. On May 16, the board responded by authorizing inter-district transfers permitting Anglo students to attend schools outside their neighborhood attendance areas for the 1974 — 1975 and 1975 — 1976 school years. 9 Executive sessions of the board met with the district superintendent and the board attorney to reevaluate the newly-enacted authorization of inter-district transfers. Counsel for the board advised the district that the court in this litigation might adversely interpret deviation from the board’s stated adherence to a neighborhood school policy, in circumstances which relieved crowding while perpetuating ethnic imbalance. On June 20, 1974, after receiving this advice, the board rescinded its- transfer resolution. Because the resolution was rescinded two months after passage, no student actually transferred from the district to a neighboring school.

Although district officials agree that double sessions are educationally undesirable, the district has maintained double sessions at Bret Harte and John Muir pending construction of new junior high schools in the southern suburban part of the district. When completed, these schools will almost certainly open as imbalanced schools, since they will serve Anglo attendance areas.

Faced with overcrowding at Pioneer High School, a southern suburban facility with a 5.7% Spanish-surnamed population, the board transferred students to Leland High School, an adjacent southern facility with a 4.1% Spanish-surnamed population. Increased ethnic balance would have resulted from board action transferring the Pioneer students to San Jose High School, a northeastern facility separated from Pioneer by the bordering Willow Glen attendance area. Although San Jose High School, with a 59.6% Spanish-surnamed student enrollment, was operating undercapacity, the district failed to transport any Pioneer students to San Jose High School.

*324 TABLE 6

Schools on Double Session — 1973

School Grades Location

Allen 2-3-4 Suburban

Almadén All elementary grades Suburban

Erickson 4-5 Suburban

Henderson 2-3 Suburban

Slmonds 2- 3-4 Suburban

Terrell 1-2 Suburban

Valley View All elementary grades Suburban

Washington 3- 4 — 5-6 Downtown

Harte 7-8-9 Suburban

Muir 7-8-9 Suburban

5. Transportation. For years, the board has authorized student transportation in the San Jose Unified School District for nonintegration purposes. In the past five years, bussing has dramatically increased in the district: the number of transported students has doubled from 5,000 in 1969 to 10,000 in 1973; the total cost of student transportation has escalated from $338,339 in 1969 to $853,380 in 1973. Indeed, bussing has become routine and widespread: all but 14 schools in the district currently employ student transportation; in the school year 1973 — 1974, 10,431 out of 36,000 students were bussed daily.

All bussing has occurred within attendance areas. No student has travelled by bus from his residential attendance area to a school in a contiguous or more distant attendance area. Thus, bussing has always been used in a manner consistent with a neighborhood school policy.

Faced with the physical and demographic makeup of the district — with a concentration of Anglo students in the southern suburban area and Spanish-surnamed students in the northern downtown core — the board could have relieved ethnic imbalance in the schools only by bussing students among attendance areas. “Two-way bussing,” in the judgment of Dr. Knight, the Task Group and the Westinghouse Learning Corporation, is a necessary ingredient in any desegregation plan.

Although aware of the impossibility of desegregation without wide-scale transportation, the board repeatedly declared itself opposed to bussing for integration. As early as 1963, the board, through resolution, assured the community of its opposition to bussing for integration; the board has not deviated from this position.

6. Bond issue elections. On February 11, 1969, a bond election was held in the district. The district divided the bond issue into two provisions (although they could have combined them into one package): Proposition A authorized the sale of bonds to finance reconstruction of Field Act schools in the northern downtown area; Proposition B authorized the sale of bonds to finance growth of the school system in the southern suburban area. The voters passed only the bonds for reconstruction of the Field Act schools.

As drafted, the bonds insured the rebuilding of each northern downtown school on site in the same locale. However, the board was under no legal compulsion to rebuild on site in the same general area.

The district scheduled another bond election for February 9, 1971. This proposal included a tax rate increase measure for operating revenues only (Proposition I), and an unrestricted 7% bond issue totaling 15 million dollars (Proposition II). Proposition II authorized 8.3 million dollars to subsidize growth and 6.7 million dollars to refinance Field Act construction since the low interest rate authorized by the 1969 measure rendered it unusable.

Prior to the 1971 election, the board issued a public statement assuring the voters that “this Board will commit itself to the continuation of the present district philosophy of the neighborhood school and transportation for specific educational and safety reasons.” Pro-bond material advised voters that if the bond issue were not passed, children presently attending northern downtown Field Act schools would necessarily be bussed to schools in the southern portion of the district. Although the bond election was held on the day of a major Los Angeles earthquake, which destroyed public hos *325 pitáis and schools, only the revenue measure passed. Funds for reconstruction of Field Act schools were obtained on April 13, 1971, when the bond market interest rate fell below 5%, enabling the district to sell the bonds authorized by the 1969 election.

The district presented the voters with another growth bond issue on September 19, 1972. Board-authorized bond material for this election promised voters that no bond money would subsidize bussing. Bond literature further threatened that failure of the bond issue would result in year-round schools as well as double sessions throughout the district. The voters rejected the bond issue.

The district attempted again to pass a growth measure on February 27, 1973. The board advised the community that failure of the bond issue would result in double sessions throughout the district and altered attendance boundaries designed to equalize student enrollment. These proposed changes, stated the board, “would affect family living patterns of most people in the district.” .The voters rejected the bond authorization once more.

7. QUEST. By resolution adopted January, 1963, the board called for a comprehensive study of racial and ethnic segregation in the school system. The resolution lay dormant for six years. In December, 1969, in response to an official state notification of racial imbalance in the San Jose schools, the board reactivated the proposed study by creating the Quality Urban Education Study Team (QUEST). The role of this board-appointed citizens’ group was limited to investigating imbalance and formulating corrective proposals; ultimate approval and implementation authority remained with the board.

Although integration necessarily requires transportation across attendance areas, the board repeatedly discouraged QUEST from including mandatory bussing in any integration proposal. Following several board meetings, the board issued a series of public statements against mass bussing. Faced with public pressure, the board opened the membership of QUEST to all concerned citizens; broad-based representation of the community resulted in a reconstituted, antibussing study group. The board ultimately modified its charge to QUEST relegating integration from the “primary objective” of the school district to a “long term . . . priority.”

In its final majority report, QUEST proposed to eliminate segregation through (1) magnet schools, (2) open enrollment policies, and (3) voluntary bussing. By 1973, the board had adopted neither the QUEST majority proposals nor the minority committee suggestion of educational parks.

8. Faculty and staff assignment. The district assigned most Spanish-surnamed teachers to predominantly Spanish-surnamed schools. Of the 1,531 classroom teachers employed by the district in 1973, 89 (5%) were Spanish-surnamed; of these 89 teachers, 54 (over 60%) were assigned to 16 downtown schools. Of the 189 administrators, counselors and resource personnel employed by the district in 1973, 20 were Spanish-surnamed; of these 20, 18 (90%) were assigned to northern downtown schools.

Dr. Knight testified that the board had consciously assigned Spanish-surnamed faculty members to northern schools based on perceived educational values. However, the record reflects that Spanish-surnamed teachers served in both northern and southern schools. Plaintiffs’ witness, Peter Mesa, agreed with Dr. Knight that there was educational value in exposing Spanish-surnamed children to Spanish-surnamed teachers in the classroom. The highest percentage of Spanish-surnamed faculty at any one time in one school was approximately 26%; another school had a 24% Spanish-surnamed teaching staff; in all other schools, the percentage of Spanish-surnamed faculty remained below 15%. On these statistics, the schools cannot be characterized as identified by segregated faculty.

9. Neighborhood school policy. The district has the authority to establish *326 and alter attendance boundaries; it approves all boundary lines. The district policy has been, and is, to require each student in the district to attend the school servicing the residence of the student’s parents. The neighborhood school policy is designed for the convenience and safety of the largest number of children; the school attempts to locate schools where it finds the children, so that students live “within walking distance of the school.” Dr. Knight defined a neighborhood school attendance area as “an area that would as nearly as possible serve as many children in such a way that as many of them as possible could walk to school.” ■

The neighborhood attendance policy is mandatory. The district has never permitted open enrollment.

Between 1963 and 1973, the district was aware that racial enrollment in each school mirrors the ethnic composition of its surrounding neighborhood when site location and construction practices were coupled with a mandatory neighborhood school attendance policy. Recognizing the correlation between residence patterns and school enrollment, the board has known that in designating attendance boundaries according to a neighborhood school policy, it has simultaneously created racially imbalanced schools. As board member McCreath testified:

Q. Now, is it true that when you are dealing with an area where the population in that area is predominantly minority, that the so-called neighborhood school policy has the effect of creating segregated schools?

A. Yes.

Q. And has the Board, since the time you have been on the Board been aware of this result of drawing the attendance district boundary lines in accordance with the neighborhood school policy?

A. Yes.

In formulating policies and procedures for facility construction or reconstruction, the district has never deviated from the neighborhood school policy. The district has never changed attendance boundaries or reassigned students to effect integration in this district.

Since 1965, the board has issued and reaffirmed a policy to establish attendance areas which would “avoid the ultimate result of segregation of pupils by ethnic or racial segregation.” Dr. Knight testified that in his view the existing attendance boundaries were too rigid and have tended to create ethnic barriers with resulting racial isolation in the school system. Dr. Knight concluded that the district could not feasibly “keep a neighborhood school policy in its pure form and at the same time have integration.”

On November 9, 1970, the board in a formal policy statement reaffirmed its commitment to the district philosophy of the neighborhood school.

C. District’s Failure to Desegregate in Violation of Board Policy and State Statutory Requirements

1. Board policy. As early as 1962, the board recognized the need for integration. In a 1963 resolution, the board officially acknowledged the existence of ethnic imbalance in the school system and resolved to undertake a comprehensive study designed to overcome “the evils produced by such school segregation.” In 1965, the board pledged to consider segregatory patterns in establishing attendance area boundaries.

Plaintiffs argue that the following board actions violated the board’s own stated policy: (1) delay in commencing a meaningful study of segregation; (2) interference with QUEST; (3) failure to adopt and implement QUEST recommendations; (4) adherence to mandatory neighborhood school • attendance areas without bussing; (5) site selection and designation of boundaries for new schools; (6) drafting bond material for replacement of Field Act schools; and (7) site selection and attendance area designation for reconstructed Field Act schools. Plaintiffs contend that, while *327 this pattern of conduct alone demonstrates compelling evidence of segregative intent, it provides conclusive evidence of segregative intent when contrasted to the stated board policy and unfulfilled board promises.

2. State policy. Title 5 of the California Administrative Code, defining racial and ethnic factors in school attendance practices, was law in California from 1962 to 1972. On October 23, 1962, the State Board of Education first filed administrative guidelines outlining the obligations of school districts in establishing attendance areas. The regulations mandated school boards to consider the ethnic composition of immediate and peripheral areas, to evaluate the impact of proposed boundary designation on ethnic balance, to formulate alternative plans, and to avoid, “as far as practicable” attendance areas which establish or maintain ethnic segregation. (Article 1.5 of Title 5 of the California Administrative Code, § 2010 filed October 23, 1962, effective 30 days thereafter [Register 62, Number 22].)

Section 2010, added to Title 5 in 1962, did not specifically legislate attendance practices or site selection. However, on February 23, 1963, the State Board amended the 1962 provisions to require school districts to analyze attendance practices, as well as attendance areas in attempting to eliminate segregation. (Cal.Admin.Code Title 5, §§ 2010, 2011, amendment filed Feb. 20, 1963, effective 30 days thereafter [Register 63, No. 3].) These provisions were not amended until 1969.

By amendments filed February 21, 1969, effective 30 days thereafter, the State Board of Education:

(a) expanded coverage to include school sites, as well as school attendance areas and school attendance practices;

(b) changed the word “segregation” to the word “imbalance”;

(c) assigned “high priority” to the prevention and elimination of imbalance in all decisions relating to school sites, school attendance areas and school attendance practices;

(d) codified the existing administrative practice requiring school districts to conduct racial/ethnic surveys and compile data recording racial and ethnic makeup of grade levels and individual schools;

(e) defined “imbalance” as a school in which one or more racial or ethnic groups differed by more than 15 percentage points from the total percentage of that group within the entire school system; and

(f) required districts to formulate and evaluate corrective plans for imbalanced institutions, assigning high priority to elimination of racial or ethnic imbalance, and analyzing “feasibility factors,” including existing demography, “traditional factors used in site selection, boundary determination, and school organization by grade level,” and “the effect alternatives on the educational program.” See Calif.Admin.Code, Title 5, §§ 2010, 2011, amendment filed February 21, 1969, effective 30 days thereafter [Register 69, No. 8]. These sections were renumbered without substantive change to sections 14020 and 14021 respectively by amendment filed January 14, 1970, effective 30 days thereafter [Register 70, No. 3].

On February 11, 1970, a court ordered desegregation of the Los Angeles Unified School District, relying in part on Title 5; one month later, on March 13, 1970, the State Board of Education issued urgency legislation repealing sections 14020 and 14021 of Title 5. [Register 70, No. 11, filed March 13, 1970, effective immediately as an emergency.] However, a Sacramento superior court subsequently determined that there existed no emergency and thus no justification for repeal without advance public notice. Consequently, the State Board of Education reinstated sections 14020 and 14021 on June 25, 1970 [Register 70, No. 26].

In 1971, Assemblyman Bagley introduced AB 724, which became law effective March 4, 1972. This legislation, which codified and revised the administrative guidelines as California Educa *328 tion Code §§ 5002, 5003, amended the Title 5 provisions to:

(1) eliminate the 15% rule and substitute “significant” difference from the district wide percentage as the governing standard for imbalance;

(2) require affirmative plans and implementation schedules to eliminate imbalance and vest approval power in the State Department of Education;

(3) order the State Board of Education to adopt rules and regulations for implementation of code sections. (Calif.Stats. 1971, Ch. 1765.)

On November 7, 1972, the California electorate passed Proposition 21, which repealed the Education Code sections and the Title 5 provisions. On January-15, 1975, the California Supreme Court upheld the voters’ authority to repeal Title 5 provisions, noting that the administrative regulations were matters of state policy not required by any constitutional or legal mandate; the Court held that the provisions of Proposition 21 repealing Title 5 were severable from the unconstitutional portions of the referendum, which sought to preclude school districts from making, assignments of students required by law.

As memorialized in correspondence and reports between the district and the State Department of Education, B.I.R., the board understood the statutory obligation to conduct studies and formulate alternative plans designed toward reducing ethnic imbalance in the school system. School district staff members presented the board with a report prepared on October 23, 1971, entitled “Preliminary Decisions Regarding Integration.” This report set forth a series of proposed decisions suggested for developing concrete policies for integration. The report recommended that planning be placed so that district-wide integration be achieved no later than the 1972-1973 or 1973 — 1974 school years. The board has never acted on this report.

Although the board has been aware of Title 5 requirements and has realized that adherence to a neighborhood school policy presents an obstacle to desegregation, the board has never directed its staff to formulate alternative criteria for student assignment or to prepare a plan for district-wide integration.

The district represented to the state in its progress reports that mandatory closing of Field Act schools would afford an opportunity to change attendance criteria and achieve progress towards desegregation; implicit in these representations was a suggestion that the board would take advantage of the opportunity. When the opportunity did arise in 1971, the board elected to rebuild the schools on the existing sites with the existing attendance areas, resulting in “very little change” in ethnic imbalance.

The district further advised the State Department of Education of progress toward integration through the QUEST study of racial imbalance and recommendations for school desegregation. After establishing QUEST, the board limited the scope of its inquiry by eliminating mandatory mass bussing from consideration as an integration measure. The final QUEST report included three proposals: magnet schools, open enrollment, and voluntary bussing. No action was taken on any QUEST proposal.

Under Title 5, school districts were required to consider the impact of site selection on ethnic composition of existing schools in adjacent areas. In selecting sites for new facilities in the Anglo suburban area, the district has not applied the state’s checklist for improved ethnic balance.

Although the board knew of the resources available at B.I.R., the district never requested technical assistance in drafting an integration plan. B.I.R. also scheduled special informational seminars for school districts engaged in desegregation litigation; - no board member attended.

V. DISCUSSION

In Keyes v. School District No. 1, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973), the Supreme Court held that *329 the desegregation mandate of Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), is triggered only by a finding of intentional state discriminatory action. Affirming the dichotomy between actionable de jure segregation and constitutionally permissible de facto imbalance, the Supreme Court stated: “We emphasize that the differentiating factor between de jure segregation and so-called de facto segregation ... is purpose or intent to segregate.” 413 U.S. at 208, 93 S.Ct. at 2697. In defining the quantum of state action necessary to sustain a finding of de jure discrimination, Keyes set's forth three requirements: (1) the school system must in fact be racially imbalanced; (2) state officials must have created or maintained ethnic imbalance; and (3) in creating or maintaining ethnic imbalance, state officials must have acted with segregative intent.

Analysis of the San Jose Unified School District within this framework establishes that two of these three elements are satisfied. As the State Department of Education has found, as plaintiffs urge, and as defendants readily concede, the San Jose school system is characterized by marked racial imbalance. 10 Moreover, while school officials did not create the imbalance, state action has perpetuated existing ethnic composition: by rebuilding the Field Act schools on site, and by adopting a neighborhood school policy to govern school assignment in a racially segregated community, the school board has maintained the existing imbalance. This litigation focuses therefore on the third element of Keyes; the central inquiry of this lawsuit is whether the numerous school board actions detailed above were motivated by segregative intent.

In the wake of Keyes, two lines of judicial authority have developed on the definition of segregative intent in school litigation. Several courts have held that a presumption of segregative intent arises once plaintiffs establish that “school authorities have engaged in acts or omissions, the natural, probable and foreseeable consequence of which is to bring about or maintain segregation.” United States v. School District of Omaha, 521 F.2d 530 (8th Cir. 1975); Hart v. Community School Board of Education, New York School District # 21, 512 F.2d 37 (2d Cir. 1975); Oliver v. Michigan State Board of Education, 508 F.2d 178 (6th Cir. 1974); Cisneros v. Corpus Christi Independent School District, 467 F.2d 142 (5th Cir. 1972) (pre-Xeyes). See also, Morgan v. Kerrigan, 509 F.2d 580, 588-589 (1st Cir. 1974), aff’g, 379 F.Supp. 410 (D.Mass.1974).

Other courts have reasoned that intent includes the concept of “purpose” within its definition. Higgins v. Board of Education, City of Grand Rapids, 508 F.2d 779, 793 (6th Cir. 1974) (there is an apparent conflict between Higgins and Oliver, both Sixth Circuit cases); Husbands v. Pennsylvania, 395 F.Supp. 1107 (E.D.Pa.1975).

Although a pre-Keyes opinion contains the language of “foreseeable consequences,” the Ninth Circuit’s post-Xeyes decisions have adopted the “purpose” rather than the “foreseeable” standard of unconstitutional segregation. In Kelly v. Guinn, 456 F.2d 100 (9th Cir. 1972), decided prior to Keyes, the court cautioned that “[w]e do not mean to suggest that we would agree with plaintiffs’ assumption that the Constitution is not violated by implementation of a ‘neighborhood school’ policy in a school district having racially segregated residential patterns when the foreseeable result [is] segregation of the races in the district’s schools. That question remains open.” 456 F.2d at 106 n. 7. Even in the same opinion, however, the court analyzed the “school district’s purpose in constructing, *330 renovating, and abandoning schools . ” 456 F.2d at 108 (emphasis added).

After Keyes, the Ninth Circuit clarified the appropriate standard for unconstitutional school board action. In Soria v. Oxnard School District Board of Trustees, 488 F.2d 579 (9th Cir. 1973), the district court granted summary judgment against the school board on a theory that “the School Board’s accountability for its segregated schools resulting as a natural, foreseeable consequence of its acts and omissions remains open for this court’s consideration, regardless of any discriminatory motives behind such acts.” 488 F.2d at 585. The trial court expressly rejected a defense based on lack of discriminatory segregative motivation, ruling that “the Board’s contention that it had never intentionally or purposefully segregated its school system did not raise a disputed question of material fact for purposes of summary judgment.” Id. Reversing, the Ninth Circuit, citing Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971), cautioned against “blaming the school district for other factors which may have contributed to some extent to the composition of schools,” and held that the school board could not be held liable for racial imbalance in the school system absent a finding of segregative intent. The case was remanded to the district court for proof on the school board’s claims that “only legitimate educational interests motivated its actions, and . it had [no] intent to create or maintain segregated schools.” 488 F.2d at 587.

Following Soria, the Ninth Circuit reaffirmed the “purposeful” standard of segregative intent. In Johnson v. San Francisco Unified School District, 500 F.2d 349 (9th Cir. 1974), the Ninth Circuit remanded a San Francisco school desegregation case for a finding on segregative intent, stating:

This court, in Soria v. Oxnard School District Board of Trustees, 488 F.2d 579 (9th Cir. 1973), has construed Keyes as requiring for any finding of unconstitutional segregation a “determination that the school authorities had intentionally discriminated against minority students by practicing a deliberate policy of racial segregation.”

500 F.2d 349, 351 (emphasis added).

Thus, under the Ninth Circuit formulation of the standard for unconstitutional de jure segregation, the school board is precluded only from practicing a purposeful policy of racial separation in the school system; the district is under no affirmative duty to improve racial balance in the schools. However, where the school board claims that actions which perpetuated racial imbalance were motivated by proper educational concerns, it is the function of the trial court to scrutinize closely the school board decision-making process to assure that false or facile justifications do not mask purposeful discrimination. We now examine the policies and actions of the San Jose Unified School District, to determine whether they indicate unconstitutional segregative intent under the Ninth Circuit standard discussed above.

A. Site Selection and School Construction

Plaintiffs stress that of nine new facilities constructed since 1965, all opened as imbalanced schools. The school board constructed all of these facilities in the southern suburban portion of the district to meet the growing suburban student population. In erecting schools where the students were located, the school board followed its neighborhood school policy.

Plaintiffs have provided no evidence that in constructing new southern schools, the school board gerrymandered or otherwise tampered with school attendance boundaries. The record discloses that in selecting sites for southern schools, the board failed to consider the state’s checklist for evaluation of attendance areas; however, consideration of the state’s ethnic criteria would clearly *331 have been futile, since adherence to a neighborhood school policy in the southern suburban portion of the district would necessarily result in racially imbalanced schools. Thus, the issue raised by site selection, as well as other school board actions, is whether the neighborhood school system is itself constitutional when engrafted on to racially segregated neighborhoods. This question will be discussed below.

B. Replacement of Field Act Schools

In rebuilding the Field Act schools, the board adhered to the neighborhood school policy and rebuilt virtually all the schools on their original sites. The two facilities, Olinder and Almadén, which were constructed on different sites, were built within the existing attendance areas.

In electing to reconstruct the northern schools within the original attendance areas, the board passed up opportunities to improve ethnic balance in the school system. For example, alternatives to rebuilding Lincoln Glen and Washington existed which would have ameliorated imbalance. From a policy standpoint, the board’s action may be condemned. The court cannot, however, say that in applying the neighborhood school policy and rebuilding the schools within the existing area the board violated the Constitution: to hold that such action constitutes segregative conduct would be to cast on the school board an affirmative duty to improve ethnic balance; this the Ninth Circuit has refused to do. Soria v. Oxnard School District Board of Trustees, supra; Johnson v. San Francisco Unified School District, supra.

C. Closure of Schools

Plaintiffs argue that discriminatory intent is shown by a series of board actions in closing certain schools, combining attendance areas, and dispersing students to adjacent facilities. As detailed above, these actions include the closures of Longfellow, Bascom, College Park, Belden, Jefferson, Wilson and Roosevelt.

In 1971, the board closed Longfellow Elementary School and combined its attendance area with Hester. A comparison of the racial composition of adjacent attendance areas shows that from the standpoint of minimizing ethnic imbalance, combination with Hester was the best available alternative within the constraints of the neighborhood school policy. Similarly, when the school board closed Bascom Elementary School and assigned its students to Trace, it selected the best ethnic alternative available under the neighborhood school policy from the adjacent facilities.

In 1973, the district closed College Park, Belden and Jefferson schools and consolidated their attendance areas with Bachrodt. A comparison of adjacent attendance areas shows that in each case Bachrodt was the best choice from an ethnic standpoint.

The closures of Wilson and Roosevelt junior high schools present more complicated fact situations. After Wilson was closed, its attendance area was dispersed among the three adjacent junior high school attendance areas, two to the north and one to the south, and one high school attendance area to the north. In selecting receiving schools, the district followed its neighborhood school policy. In each case, the students from Wilson found themselves at a school where ethnic imbalance was lower after their arrival than it had been at Wilson. Naturally, the number and percentage of Spanish-surnamed students at each of the receiving schools increased; the most imbalanced junior high school was, however, eliminated. Although the changes improved imbalance from the viewpoint of the Wilson students, they aggravated it from the viewpoint of students at receiving schools. On the whole, however, the court finds that the changes did not significantly aggravate imbalance in the district.

The elimination of the Roosevelt school attendance areas and reassign *332 ment of students had no significant impact on ethnic imbalance. Moreover, the changes were conformity with a neighborhood school policy; the alternative options available to the board each required deviation from the neighborhood school policy.

D. Use of Portables and Double Sessions

In allocating portable classrooms to existing facilities, the school board has used mobile units in accordance with its neighborhood school policy, to allow students to remain in their residential attendance area.

Plaintiffs stress the fact that the board instituted educationally undesirable double sessions in the southern portion of the district rather than transporting students to the undercapacity northern schools. While one may disagree with this choice, it demonstrates not racial animus but only strict adherence to the neighborhood school policy. The only deviation from the neighborhood school policy was the board’s authorization for inter-district transfers from the congested Bret Harte and Muir junior high schools. Since this short-lived resolution was soon rescinded, no student transferred under the proposed plan.

E. Transportation

Plaintiffs emphasize that while the district uses extensive student transportation, the board has never implemented — or even considered — bussing for integrative purposes. This evidence shows only that the district has failed to achieve a perfect “walking distance” system. However, all student transportation is within attendance areas; thus, the district’s bussing policy is consistent with its neighborhood school policy. The record discloses no evidence that an equal amount of student transportation could have improved racial balance in the school system. This is a far cry from the once-common case where minority students were bussed past an undercapacity Anglo facility to attend a racially isolated school. Milliken v. Bradley, 418 U.S. 717, 725, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974).

F. Bond Issue Elections

As discussed above, reconstruction of Field Act facilities within existing attendance areas did not violate the Constitution. Thus, the fact that the provisions of the bond issue submitted at the February, 1969, election limited funds to rebuilding on site cannot constitute a constitutional violation if the reconstruction itself was permissible.

After passage of the 1969 bond election, decline in interest rates rendered the bonds unsalable and necessitated a second bond election for reconstruction of Field Act reconstruction. Pro-bond literature in the 1971 election was factually accurate in its representation that failure to rebuild northern schools would result in northern students transferring to southern schools. Some pro-bond voters may have been motivated by a fear or dislike of Spanish-surnamed students in southern facilities; however, they may also have been motivated by a fear of congestion or a multitude of other considerations. It is difficult to speculate on the voters’ reactions to pro-bond literature; at any rate, the bonds did not pass. The court finds that the factually accurate statements of the board do not establish segregative intent.

The repeated assurances during various bond elections that no funds would subsidize bussing may have assured voters who feared forced integration. However, if there was no affirmative constitutional obligation to bus, these pro-bond representations cannot form the basis for a finding of segregative intent.

G. QUEST

The record shows that the board was less than supportive of QUEST. However, the board had no affirmative duty to integrate; thus, the failure to adopt or implement an integration plan does not show segregative intent. Nor is purposeful segregation shown by the board’s discouraging QUEST from focusing on bussing as a cornerstone to an integra *333 tion plan; as stated above, extensive bussing would violate — if not destroy— the district’s long standing neighborhood school policy.

H. Faculty and Staff Assignment

As noted in the presentation of facts, no school in the district can be identified by its segregated faculty. The board did intentionally assign Spanish-surnamed faculty members to predominantly Spanish-surnamed schools; the board also assigned Spanish-surnamed teachers to Anglo schools.

Courts have frequently condemned segregation of faculty. Segregation in the assignment of faculty and staff “ . . . [is] among the most important indicia of a segregated system.” Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 18, 91 S.Ct. 1267, 1277, 28 L.Ed.2d 554 (1971). Discrimination in the assignment of faculty and staff violates the Constitution independently of the segregation of students. United States v. Montgomery County Bd. of Educ., 395 U.S. 225, 89 S.Ct. 1670, 23 L.Ed.2d 263 (1969); Bradley v. School Bd. of City of Richmond, 382 U.S. 103, 86 S.Ct. 224, 15 L.Ed.2d 187 (1965); Davis v. School District of Pontiac, 309 F.Supp. 734 (E.D.Mich.1970) aff’d, 443 F.2d 573 (6th Cir.), cert. denied, 404 U.S. 913, 92 S.Ct. 233, 30 L.Ed.2d 186 (1971). In the context of discrimination against Blacks, assignment of teachers on the basis of skin color is justly condemned since no educational rationale can be advanced to support racial isolation among faculty. In a suit such as this involving discrimination against Spanish-surnamed Americans, the problem is complicated by language differences. Students for whom English is a second language may or may not need teachers who share their primary language. Proper analysis of this problem requires information on the percentage of Spanish-surnamed students who benefit from bilingual teachers and the extent to which only Spanish-surnamed instructors may assist the special problems of bilingual students. Many Spanish-surnamed students are perfectly fluent in English. Moreover, Spanishsurnamed teachers have no monopoly on speaking Spanish. Unfortunately, no party produced evidence on these questions at trial.

Considering that the percentage of minority teachers in any school is small, that Spanish-speaking teachers may benefit Spanish-speaking students, and that Spanish-surnamed teachers have been assigned positions at Anglo schools, the court does not find that district practices violate the Constitution.

The court further finds that assignment of minority teachers in this case does not in any manner contribute to ethnic imbalance of students in the schools. Even if the assignment practices in the district were unconstitutional, the appropriate remedy would relate solely to faculty assignment and not to student attendance policies. If the assignment of students is otherwise constitutional, it should not be disturbed simply because minority teachers have been assigned to those schools which have constitutionally permissible percentages of minority students. Higgins v. Board of Education of City of Grand Rapids, supra, at 787.

I. Failure to Integrate Despite State and Board Integration Policies

Plaintiffs argue that segregative intent may be found from the fact that the board adopted a public policy favoring integration but then failed to achieve its stated goal. Throughout the time in question, the board adhered to the neighborhood school policy. Complete integration and a neighborhood school policy are incompatible in the San Jose Unified School District.

Plaintiffs’ argument is somewhat unfair to a board facing numerous complex decisions. Had the board failed to state that integration was a worthy objective, plaintiffs could not attack it for unfulfilled promises. Defendants should not be hanged by their own rope simply because they recognized the racial imbalance in the school system, but adhered *334 to the neighborhood school policy rather than taking affirmative action to integrate.

Plaintiffs also urge the court to find segregative intent from the defendants’ failure to integrate despite an alleged state-imposed duty. It is clear that the district failed to implement integration proposals. However, the record is ambiguous both as to the nature of the duty imposed and the board’s response. As shown above, the state integration guidelines did not impose an absolute affirmative obligation to integrate the district; the guidelines were qualified by other educational considerations. Moreover, the state duty was in constant flux: between 1969 and 1972, the guidelines were strengthened, repealed, ordered reinstated, reinstated, modified by legislative action and ultimately repealed by the electorate. Throughout this period, the district worked with the QUEST committee, the Building Master Plan Task Group and state consultants in an attempt to discharge its obligations. The court does not find evidence of segregative intent from the district’s failure to integrate during the years the state guidelines were in effect.

J. Neighborhood School Policy

The court finds that the district has consistently adhered to a neighborhood school policy. The board has applied this policy neutrally: the record discloses no attempts to gerrymander attendance boundaries or otherwise manipulate attendance areas to lock in minorities or freeze segregated school patterns.

Plaintiffs repeatedly argue that defendants failed to take advantage of alternatives that would have reduced ethnic imbalance. In discussing the facts, the court has noted the rejected alternatives. The court may disagree with the policy of the board in pursuing other educational goals over improved ethnic balance. If, however, neutral adherence to a neighborhood school policy is constitutional, this court has no authority to intervene and order integration.

The critical issue in this litigation is thus whether a school board may apply a neighborhood school policy. Plaintiffs cite two cases in support of their argument that a neighborhood school policy is unconstitutional. In Brewer v. School Board of City of Norfolk, Virginia, 397 F.2d 37 (4th Cir. 1968), the court held that school assignment by neighborhood school policy was impermissible, where Norfolk had operated a dual school system in the past and, thus, had a duty to remedy the effects of past discrimination. Plaintiffs in Brewer also charged that the school board gerrymandered attendance areas for racial purposes. By the time of trial, the school system used attendance zones with several schools in a zone; students had freedom of choice to attend any school within the attendance area. The “geographical zoning plan” had been adopted without redrawing existing boundary lines, which had been used in the past for a modified freedom of choice plan. Brewer is distinguishable from the present case in that the case involved a statutory dual school system. In striking down the Norfolk plan, the court did observe that “[assignment of pupils to neighborhood schools is a sound concept, but it cannot be approved if residence in a neighborhood is denied to Negro pupils solely on the ground of color.” 397 F.2d at 42.

The second case cited by plaintiffs is Spangler v. Pasadena City Board of Education, 311 F.Supp. 501 (D.C.Cal.1970), in which the district court held that a neighborhood school policy is unconstitutional when it results in imbalanced schools. The court ruled that “[ujnder the Fourteenth Amendment a public school body has an obligation to act affirmatively to promote integration.” 311 F.Supp. at 521. This decision, relying in part on Brewer, was handed down three years before the Supreme Court spelled out the requirement of segregative intent in Keyes. Spangler was never reviewed by the Ninth Circuit because the school board acquiesced in the court’s order. 427 F.2d 1352 (9th Cir. 1975). In

*335 view of these facts, this court is not disposed to follow Spangler.

Several cases have approved neutral adherence to a neighborhood school policy. In Oliver v. Michigan Board of Education, supra, 508 F.2d 178, 182 (6th Cir. 1974), the Sixth Circuit held that a school board could rebut the presumption of segregative intent by establishing that its conduct was “a consistent and resolute application of racially, neutral policies.” 508 F.2d at 182. This same court expressly approved adherence to a racially neutral neighborhood school policy even though it resulted in ethnic imbalance in the school system. Higgins v. Board of Education of City of Grand Rapids, supra, 508 F.2d 779, 783, 791 (6th Cir. 1974).

The Ninth Circuit has commented on the neighborhood school policy in Johnson v. San Francisco Unified School District, supra:

The Court in Keyes specifically reserved the issue of “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.” 413 U.S. at 212, 93 S.Ct. at 2699.

We understand the reserved question to be whether it is sufficient to show intentional discriminatory action by someone other than the school board resulting in racially divided neighborhoods, to which the school board applies a “neutral” neighborhood school policy. Cf. Kelly v. Guinn, 456 F.2d 99, 106 n. 7 (9th Cir. 1972). Engrafting a neighborhood school policy onto such involuntary neighborhoods may be sufficient ratification of the illicit intent of others to preclude the necessity for showing a purpose by the school board itself to segregate.

500 F.2d at 351, n. 1.

This language indicates that a neighborhood school policy is constitutionally suspect only where the neighborhoods which define school attendance areas are themselves products of official state discriminatory action. In the present case, plaintiffs produced no evidence of state-approved restrictive covenants, discriminatory zoning, or other racially-motivated state action shaping the demographic and ethnic makeup of the San Jose Unified School District. Since the school board applied the neighborhood school policy neutrally, it follows that the record is barren of evidence of state discriminatory action on the part of the school board or any other political entity.

From a policy standpoint, much can be said for the neighborhood school system. In this day of ever-increasing centralization, the average citizen has less and less control over his life. Decisions by distant bureaucrats appear unresponsive to a citizen’s wishes. Transporting students to schools far from home only increases the feeling that the parent is powerless to affect school policy, or to participate in the educational process. Neighborhood schools allow parental involvement in the life of the school. Additionally, time and safety considerations both support the use of neighborhood schools, especially for younger children. This court is not attempting to provide the definitive word on education; these comments are simply to indicate that the neighborhood school system is at the least a reasonable approach to education.

VI. CONCLUSION

The court finds that defendants, who consistently adhered to a neighborhood school policy, have never acted with segregative intent. The schools are, therefore, not unconstitutionally segregated; and this court has no authority to order integration. As the Supreme Court stated in Swann v. Charlotte-Mecklenburg Board of Education, supra:

It is important to remember that judicial powers may be exercised only on the basis of a constitutional violation. Remedial judicial authority does not put judges automatically in the shoes of school authorities whose powers are plenary. Judicial authority enters only when local authority defaults.

402 U.S. at 16, 91 S.Ct. at 1276.

*336 We do not, however, intend to brand the current allocation of students with a judicial imprimatur, or to discourage the school board from attempting to improve racial balance in the school system. Nothing decided here is intended to inhibit the school board from seeking creative solutions to the problem of ethnic imbalance or to deter the community from supporting such efforts.

This opinion shall constitute the findings of fact and conclusions of law under Rule 52 of the Federal Rules of Civil Procedure.

. Lincoln Glen was not mentioned in the body of the complaint, but was included in the prayer and in the Application for Temporary Restraining Order and the Order to Show Cause.

. Longfellow and Edison operated on the same campus, which is no longer the case. Plaintiffs have not pressed their Edison claim, Edison serving the entire school district as a continuation/opportunity school.

. Because construction bids had already been received on Hester and Lowell elementary schools, this court found that the school district would suffer substantial harm if rebuilding of them were halted. College Park was inadvertently omitted from the Temporary Restraining Order.

. This ignores a narrow corridor extending north to the bay in which there is no resident population.

. “Safety” is determined by the particular traffic hazards, availability of sidewalks, distances to school, and the age and grade level of the students.

. School buildings constructed under the supervision of the Department of General Services in accordance with approved plans were exempted from this requirement.

. Before Bascom was built, Trace served the Bascom area.

. The Bachrodt area had previously been served by College Park, and Jefferson had at one time served a portion of the Belden attendance area.

. No evidence was introduced at trial regarding the comparative distances between the adjacent southern schools and the undercapacity northern facilities.

. There is no evidence, however, that the schools are imbalanced in any other respect, i. e., that there is an inequitable distribution of any educational resource — trained faculty, recreational or housing facilities, equipment, or textbooks — disfavoring students at northern schools.