Lipscomb v. Wise, 399 F. Supp. 782 (N.D. Tex. 1975)
March 25, 1975
The Honorable Wes WISE, Mayor of the City of Dallas, et al.
United States District Court, N. D. Texas, Dallas Division.
*783 James A. Johnston, Johnston & Dixon, Sylvia M. Demarest, Dallas Legal Services, Edward B. Cloutman, III, Mullinax, Wells, Mauzy & Baab, Walter L. Irvin, Hudson & Irvin, Thomas W. Pauken, Dallas, Tex., for plaintiffs.
N. Alex Bickley, City Atty., Asst. City Attys. Joseph Werner, Lois Bacon and Lee Holt, Dallas, Tex., for defendants.
Frank P. Hernandez, Dallas, Tex., for intervenor Mexican-Americans.
MAHON, District Judge.
This challenge to the present electoral system for the Dallas City Council has progressed a winding path to trial. Filed in 1971, the suit was originally dismissed for failure to state a claim upon which relief could be granted. On appeal, the Fifth Circuit remanded with clarification of what plaintiffs' burden would be in these circumstances. Lipscomb v. Jonsson, 459 F.2d 335 (5th Cir. 1972).
Plaintiffs originally purported to represent the black population residing within an identifiable area of the City of Dallas, but prior to December 9, 1974, no order certifying this cause as a class action was entered. The Court originally began to hear testimony on this matter during the week of July 15, 1974. After approximately one and a half days of testimony, the Court, on its own motion, and out of an abundance of caution, suspended testimony pending resolution of the possibility that the matter might have been one of state-wide rather than local concern, and therefore appropriate for a three-judge court determination. The case was determined to be one appropriate for a single judge to hear, and the cause was reset for hearing for the week of December 9, 1974. On the day of the trial, the Court entered an order, certifying the matter as a class action; the class consisting of all blacks residing within the corporate limits *784 of the City of Dallas. The order also denied a pending motion for leave to intervene on behalf of certain Mexican-Americans as parties plaintiff. This denial was with the specific right of the proposed intervenors to participate in any subsequent hearing that may be held concerning the manner of appropriate relief if the present election system was held to be constitutionally defective. The order provided in part that:
"It appeared that on the initial filing of this lawsuit on the 10th day of March, 1971, there were included as plaintiffs members of the Mexican-American Class and the action was on behalf of all the minority groups including the Blacks and Mexican-Americans who live within the City of Dallas. The defendants sought to elicit written interrogatories from all of the plaintiffs. Certain of the plaintiffs refused to respond to the interrogatories and upon contact by their counsel continued to refuse to respond. Included in this group were all of the Mexican-American Race originally included as plaintiffs in the lawsuit.
Upon motion duly made and hearing had, the Court dismissed from the lawsuit all of those members of the plaintiff's class who refused to respond to written interrogatories. Included within this group were all those of Mexican-American descent.
Prior to the entry of this Order the plaintiffs had amended their pleadings by their Second Amended Original Complaint and included in the Amended Complaint were both the Black and Brown citizens of Dallas. However, counsel determined that after all of the Mexican-American plaintiffs were dismissed, and having no individual Mexican-American plaintiff, that they could no longer represent the Mexican-Americans as a class and therefore urged the Court to consider this an action with the class being those Black citizens residing within the city limits of the City of Dallas.
After a portion of the case had been presented to the Court and the trial had been commenced and recessed, the intervenors filed their motion on July 16, 1974, for the purpose of intervening on behalf of the Mexican-American citizens in the City of Dallas. The testimony previously presented to this Court shows that this class represents approximately 8% of the population of the City of Dallas, the housing pattern is such that they are spread throughout the city limits of Dallas and are not a predominant factor in any concentrated census tract groups, and in view of these facts, the Court is of the opinion that to allow the intervention at this time would result in many difficulties in the management of this case as a class action and it would result in a delay that would possibly take the case beyond the time of the City Council election and the time that candidates need to file their candidacy. Moreover, the Court is of the opinion that to permit this intervention at this late hour would not assist the Court in resolving the issue of the constitutionality of the present at-large system."
After this order was entered, testimony was resumed on the question of whether the existing manner of electing members of the Dallas City Council was constitutional. On January 17, 1975, the Court found the present system of electing members to the Dallas City Council constitutionally infirm. Dallas was afforded an opportunity to come forward with an apportionment plan which would meet constitutional standards and testimony was heard on the merits of the city's proposal the week of February 4, 1975. Testimony at the remedy stage took over two days.
This suit is not the conventional one person one vote reapportionment case; rather it is an attack alleging dilution of the racial minority vote. The dilution is said to arise from operation of the atlarge *785 system of electing Dallas City Council which provides that every candidate for council run city wide, and face all the voters of the city. The Dallas City Charter requires that the city be divided into eight residential districts for City Council elections. Any person seeking election for any of the "places" on the ballot corresponding to the eight districts, must reside in the respective district. Three members, including the mayor, run without regard to any residence requirement. Voting for all eleven council seats is at-large, that is, on a city wide basis, regardless of the residency requirement. A majority of all votes cast for the councilmen, for the place for which the person is running, is required for election.
Dallas adopted its present council-manager form of government in 1931. Prior to that time it had a commission form of government, but since at least 1907, members of the city government have been elected on an at-large basis.
The number of positions on the City Council has been enlarged from time to time, the latest change being in 1968 when, by charter amendment it was increased from nine to the present eleven seats. The City Council elects from among its members two councilmen to hold the positions of Mayor Pro-Tem, and Deputy Mayor Pro-Tem. As noted above, all positions to council are elected by a vote of the entire city.
The alleged dilution of the black citizens vote was said to have its genesis in the racially segregated housing pattern present in the City of Dallas. Plaintiffs' uncontradicted evidence showed the existence of what was variously called a minority or inner city area. The characteristics of this geographical area were developed by use of United States Census demographic data and the area was described as being more or less contiguous, extending from the city's South side, north to the city's central business district, curving northwest across the business district to the city's Love Field area. Within this area are located some forty-odd standard statistical community census tracts as defined by the United States Census Bureau. Plaintiffs' experts gave testimony concerning the racial and socio-economic composition of the minority area and offered exhibits in summary. The black population of Dallas is 210,227. The evidence shows that the minority area contains a black population which is 90% of all residents in the area. This means that nine out of every ten residents are black, of all the black residents in the city of Dallas, 93% reside within this area; of all the black residents of Dallas County, 95% were shown to reside here. In addition to the distinct racial composition of the inner city area, other characteristics were shown. A less desirable rating on many economic indicators of well-being were shown to exist here. For example, housing was shown to be generally of less value and of lower quality in the inner city area. The median grade level of schooling attained by residents of this area was shown to be less than for the non-minority areas. Unemployment was higher. Median income was lower. In general, the socio-economic indicators purporting to show "quality of life" were less favorable when compared with other areas of Dallas.
The voting patterns of the inner city area in relation to the voting patterns of the rest of the city with regard to elections for City Council positions were examined. The testimony was concerned with who carried the black area and how that candidate fared in the city elections at large. Examples from five elections demonstrate the pattern which emerged from this testimony. The first two elections concern place three for the subsequent election years of 1959 and 1961. The 1959 election presented a black vs. white contest. The black candidate in *786 that election polled some 87% of the votes from the above-described inner city. His white opponent polled some 73% of the vote from the non-minority area. The result, when translated into vote totals, gave the white candidate a 65% total of all votes cast and made him the election winner. The 1961 election pitted a black against two white opponents. Here, the black candidate played the role of spoiler, forcing a run-off between his two white opponents. In the first election the black candidate garnered 81% of the black area's votes. In the white areas he gathered a scant 15%. The relative percentages left him as low man in the three-way race and eliminated him from the run-off. In 1965, the place eight election is illustrative of the pattern. This was a two-way race, black versus white. The black candidate carried the black area with 86% of the vote; the white candidate carried the white area with 75% of the vote. The total vote percentage gave the white candidate 62% making him the winner of the election. In 1969, the place ten race was three-way involving one black against two whites. The black carried the black area with 42%. The two whites together pulled about 83% of the white area's vote. No run-off was necessary however because here one of the whites received 59% of the total votes within the city. The 1971 election showed a corollary situation for place ten. Here, two black candidates ran against a single white candidate. Together the black candidates received 60% of the black area vote. The white candidate received 68% of the vote from the white areas, which was 63% of the city wide total. This, of course, made him the winner without the necessity of a run-off. The Court feels that it is this voting pattern which is the key factor to understanding the dilution present in Dallas.
A description of Dallas political life would be incomplete without reference to the non-partisan slating groups. Like so many large cities, partisan politics in the "party" sense of the word does not exist. The Republican and Democratic Parties simply do not field candidates. Instead, the politics surrounding Dallas City Council elections turn on the influence of slating groups. The slating group which has enjoyed dominance in the sense that it has been extremely successful in getting its candidates elected is the Citizens Charter Association (hereinafter C.C.A.). The C.C.A. has as its express goal the maintenance of good municipal government, specifically as embodied in the council-manager form. Testimony by one active participant in the affairs of C.C.A., an ex-president, gave insight into its functioning.
C.C.A. has enjoyed considerable success with its candidates. Testimony shows that in the eight elections held since 1959, C.C.A. endorsed candidates have won in 82% of the races involved. This history shows that of seventy-five C.C.A. endorsed candidates, sixty-four *787 triumphed at the polls. This is an 85% success rate.
The record affirmatively shows that since 1907 only two blacks have been elected to the Dallas City government. The first black to be elected was the present Mayor Pro Tem, Mr. George Allen. Mr. Allen testified for the plaintiffs in the first stage of the trial and for the City at the remedy stage. His testimony is that he was first elected to City Council in 1969, but only with C.C.A. support. This support was said to have been the result of a campaign bargain in the 1967 election to the effect that the black support for C.C.A. endorsed candidates would be linked to support for a Charter Amendment increasing the size of City Council from nine to eleven and that one of the two new seats would go to a black person. This bargain was apparently kept by all parties. The Charter Amendment was passed, the size of the City Council was increased, and Mr. Allen was supported for a position by C.C.A. and won in 1969 with 71% of the vote. His opponent in that election was also a member of the black race. The other black person to be elected to City Council was also endorsed by C.C. A. Mrs. Lucy Patterson ran in an all black contest for place eight in 1973. She had a plurality in the general election and went on to win in the run-off with 53% of the vote total.
The above-mentioned facets of the Dallas redistricting question must be viewed in an historical context. Dallas, located in the state of Texas, inherited by default some of this state's history of official race segregation. There is evidence, however, that Dallas also participated in this history of discrimination. However unpleasant it is to face that history, it cannot be avoided. Dallas' participation in that history is before the Court in large measure by stipulation of the parties. A review of those stipulations shows among other matters the following:
a. The original charter for the City of Dallas written in 1907 contained a section entitled "Segregation of the Races." This section was amended through 1952 and was carried forward as part of the charter until it was repealed in 1968. This section authorized the City Council to pass city ordinances providing for the use of distinct blocks, for the housing, for amusement, for churches, for schools, by members of the "white" and "colored" races.
b. In 1937 the City Council passed on ordinance regarding separate spaces in commercial motor vehicles for white and black passengers. A penalty was established for those who rode in spaces not designated for the race of the individual involved.
c. In 1942 the City Council adopted a resolution which enumerated the requirements which a taxi cab owner must have met before the cab would be permitted to carry Negro passengers.
d. In 1961 the City Council agreed to contract for the engaging of ambulance service and burial of Negro paupers.
e. In 1973 the City of Dallas City Council recognized by resolution that there is "unequal law enforcement, dual justice and unequal treatment for the different segments of the community and different races." There are other similar ordinances included by stipulation but these four will suffice for example. There was additional testimony from several witnesses concerning this history and the negative impact it made on the black community generally and specifically on the political life of that community.
*788 Two currents make up the equal protection argument swirling around reapportionment cases. The original stream is the one person, one vote concept. This principle of equal protection is offended when some votes carry more weight than others. Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506 (1964). One person, one vote came to Texas local governments in 1968. Avery v. Midland County, Texas, 390 U.S. 474, 88 S. Ct. 1114, 20 L. Ed. 2d 45 (1968). Dilution is the other more recent current. This concept involves election schemes conceived or operated as devices to further racial discrimination and a corresponding lesser degree of opportunity to participate in the political process. Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858, 29 L. Ed. 2d 363 (1971). The Supreme Court in recognizing that multi-member districts inherently apply the one man, one vote concept, stated that circumstances of a particular case may operate to minimize or cancel out the voting strength of racial or political elements of the voting population. Whitcomb v. Chavis, supra, citing Burns v. Richardson, 384 U.S. 73, 86 S. Ct. 1286, 16 L. Ed. 2d 376 (1966); Fortson v. Dorsey, 379 U.S. 433, 85 S. Ct. 498, 13 L. Ed. 2d 401 (1965). The Fifth Circuit panel reviewing the initial dismissal of this case on appeal outlines several factors by which plaintiffs could show lack of participation in the political process of the life of Dallas. Lipscomb v. Jonsson, 5 Cir., 459 F.2d 335 (1972). Since that appeal, the Supreme Court has given additional guidance to the circumstances which have a bearing on the opportunity to participate. In Graves v. Barnes, 343 F. Supp. 704 (W.D.Tex.1972), aff'd in part sub nom.; White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), Texas legislative reapportionment was under scrutiny. White noted that the Supreme Court had entertained claims that multi-member districts were "being used invidiously to cancel out or minimize the voting strength of racial groups . . . [and] to sustain such claims, it is not enough that the racial group allegedly discriminated against has not had legislative seats in proportion to its voting potential. The plaintiffs' burden is to produce evidence to support findings that the political processes leading to nomination and election were not equally open to participation by the group in question, that its members had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice." White, supra, 412 U.S. at 766, 93 S. Ct. at 2339, 37 L. Ed. 2d at 324. Looking at this standard, the Supreme Court observed several factors upon which the three-judge court had predicated their holding that the existing scheme was invalid. Among these factors was the history of official race discrimination which at times touched the right of Negroes to vote. Other factors included the number of blacks elected under the existing scheme and the correlation between the endorsement of the white dominated slating organizations and electoral success. White, supra, at 755, 93 S. Ct. 2332, 37 L. Ed. 2d at 324-25.
Plaintiffs' ultimate claims in the case sub judice reduce to the question of whether of not the electoral scheme for the Dallas City Council operates to minimize or cancel out the voting strength of racial elements of the voting population. This question was considered with regard to a court approved apportionment plan involving a parish school board and police juries, wherein at large elections the juries were provided in a Louisiana parish where the so-called racial minority commanded a majority of the entire parish population. The Fifth Circuit stated *789 that the proper measure for measuring dilution of minority voting strength is not population but rather access to the political process. Zimmer v. McKeithen, 485 F.2d 1297, 1303 (5th Cir. 1973) (en banc). The Court reviewed the Supreme Court decisions dealing with dilution or restriction to access and highlighted the various factors which could contribute to the showing of the fact of dilution. Included among these factors are:
1. a lack of access to the process of slating candidates;
2. the unresponsiveness of legislators to particular minority concerns;
3. a tenuous governmental policy underlying the preference for at large districting;
4. the existence of past discrimination in general which would preclude effective participation in the election system. Zimmer v. McKeithen, supra, at 1305.
The Court emphasized the Supreme Court's holding that dilution can be established upon a showing of an aggregate of factors and therefore that it is not necessary that each or all factors be proved. Zimmer v. McKeithen, supra, at 1305.
Two recent cases by the Fifth Circuit illustrate some ramifications of dilution. In Robinson v. Commissioner's Court, Anderson County, 505 F.2d 674 (5th Cir. 1974); a panel of the Fifth Circuit considered dilution in the context of manipulation of district boundaries, i. e., gerrymandering. There reapportionment of the County Commissioner's Court was considered for a Texas county wherein whites comprised approximately 75% of the population and blacks 25%. The major city in Anderson County is Palestine. The city's racial composition was 74% white and 26% black. The greatest concentration of blacks within the county was within one section of Palestine and 83% of the population of this section was black. In this section of Palestine approximately 26% of Anderson County's total black population resided. Anderson County is as are all Texas counties, divided into four precincts for the purposes of electing county commissioners. The redistricting scheme under scrutiny in Anderson County was drawn in 1969. It divided the black section of Palestine, Texas, into three separate precincts. In other words, the black community was segmented and divided for the purpose of electing County Commissioners. The panel upheld a district court finding of dilution. It was fragmentation of what was otherwise a "cohesive voting community" that was crucial to the finding of dilution. Robinson v. Commissioner's Court, Anderson County, supra, at 679. In Reese v. Dallas County, Alabama, 505 F.2d 879 (5th Cir. 1974) (en banc), the Fifth Circuit sustained a finding of dilution in an apportionment plan which did not involve race. Here the disparity was between rural and urban segments of the county's population. Dallas County, Alabama, is also divided into four districts for the purpose of electing County Commissioners. The district containing the City of Selma, Alabama, contained approximately half of the total population for the county. The three other districts divided the remainder of the population unevenly. The plan involved at large elections with a resident place requirement. In other words, for each of the four districts, a candidate for that district place on the County Commissioner's Court must reside within the district, although he was voted on at large. All voters of Dallas County, *790 Alabama, vote for all four commissioners and only one of the candidates residing in each district can be elected. With this election scheme, dilution was found with regard to the urban voters of Selma. The Selma voters were "forbidden to elect resident commissioners in proportion to their numbers. . . . Unlike residents of Selma the other voters in Dallas County [Alabama] can choose their proportional share of the representative body, from among . . . candidates who can . . . be expected to share their interests." Reese v. Dallas County, Alabama, supra, at 883. In other words, the scheme involved a minimizing or cancelling out the voting strength of a group that had an identifiable set of common interests. In Reese, the groups with an identifiable set of common interests were divided along rural and urban lines. In the instant case the landmarks of the set of common interest are racial.
From these two cases it can be seen that the fact of dilution can be present in a variety of voting schemes, even when the one person, one vote concept is adhered to. Dilution was found with single-member districts when a concentrated group with common interests was split up diffusing its strength over three districts. Robinson v. Commissioners Court, Anderson County, supra. Likewise, in Reese, with an at-large scheme, dilution occurred where candidates were required to be residents of unequally populated districts and the districts had obvious distinctions in "identifiable sets of common interest." Reese v. Dallas County, Alabama.
Based on these standards, it is clear that the present system of requiring all members of the Dallas City Council to run at-large on a city wide basis involves dilution. In this regard, the Court believes that two factors are of particular significance. These are the existence of past discrimination in general, which precludes effective participation in the electoral system and a customary lesser degree of access to the process of slating candidates than enjoyed by the white community. These factors are crucial when viewed in the historical context of the city and state.
The mere existence of a definable minority area wherein some 90% of all black residents in the city live, the Court believes, is itself a lingering effect of past official race discrimination. There appear to be other lingering effects and these include the lesser degree of opportunity available to black residents to meaningfully participate in the election process in the City of Dallas. This lesser degree of opportunity is best shown by the voting pattern I have found to exist for the city:
Black voters, that is, those residing in the inner city area, vote for black candidates, giving them at least a plurality, and usually a majority of their votes, and the white community, the non-minority voter tends not to vote for the black candidates.
This is dilution. In other words, when all members of the city council are elected at large, the significance of this pattern of blacks carrying their own areas and yet losing on a city wide basis is that black voters of Dallas do have less opportunity than do the white voters to elect councilmen of their choice. Another shadow of dilution is found in the high correlation between endorsement by the C.C.A. and victory city wide. Meaningful participation in the political process must not be a function of grace, but rather is a matter of right. Graves v. Barnes, 343 F. Supp. 704, 726 (W.D.Tex.1972), aff'd in part sub nom. White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 314 (1973).
The Fifth Circuit in Zimmer v. McKeithen, supra, noted several factors to be considered in establishing dilution. As noted above, one of those factors is responsiveness of legislators to particular minority concerns. Dallas presented *791 much evidence on the city's present responsiveness. It is the Court's view that if present responsiveness of the city to the particular interests of the black community (or for that matter the brown or white communities), were the sole issue which determined dilution, then I have little doubt that there would be no finding of invidious discrimination. There has been ample demonstration of the fact that Dallas is acting for the needs of all its citizens parks, street services, police and fire protection, transportation, equal employment opportunity, fair housing and community relations. In these areas and more I find the city to be acting in a responsible manner in a responsive manner. This present responsiveness, however, is not enough to justify the present exclusive at-large voting plan when weighed against the other factors which I have found.
On January 17, 1975, I made preliminary oral findings of dilution which rendered the exclusively at-large system of voting unconstitutional. After being given time to consider its legislative responsibilities, the city offered its plan and plaintiffs in response, offered two proposals which they felt would be an effective remedy for the dilution which the Court had previously found. The Mexican-American intervenors as permitted by the Court's order of December 7, 1974, participated in this hearing and elicited testimony as to the effect that the city's plan and the plaintiffs' two plans would have on the Mexican-American voting population of Dallas. Additionally, evidence was heard on behalf of some dissenting members of plaintiffs' class who supported the general concept of an exclusive single-member district plan, but disagreed with the implementation of this concept as exemplified by plaintiffs' two proposals.
The election scheme presented by Dallas is a combination, single-member district and at large voting plan. It provides for eight single-member districts and three at-large positions including the mayor (hereinafter sometimes called "city's plan" or "eight/three plan"). The eight council persons elected from the respective eight single-member districts must reside within the district they represent and are to be elected by a majority vote of the residents of their district. The three at large seats have no residency requirement and are elected by a majority vote of all votes within the entire city. One of the at-large council seats is designated as mayor. The district boundaries are essentially the same as the resident district boundaries under the exclusive at-large voting plan. Plaintiffs presented two alternative plans. Their Plan D provides for ten single-member districts and a mayor who would be elected at large. Each council person would reside in and be elected by a majority vote of his respective district. The mayor would have no residency restriction (hereinafter "ten/one plan"). Plaintiffs' alternate plan provides for each of the eleven council members to be elected from a district and would have the mayor selected by the council members themselves from one of their number (hereinafter "eleven/zero plan").
The Court considered each of the three plans presented, with the spirit of *792 the Supreme Court's mandate in Chapman v. Meier, 420 U.S. 1, 95 S. Ct. 751, 42 L. Ed. 2d 766 (1975) in mind. That is, that reapportionment is primarily the responsibility of state legislative bodies and not the federal courts. In the case at bar, an existing method of electing city council was found to be constitutionally defective. That plan was declared invalid and this Court then gave the City of Dallas an opportunity to perform its duty to enact a constitutionally acceptable plan. I find that it has met that duty in enacting the eight/three plan of electing council members.
The initial question encountered when considering the city's plan, is whether or not any plan providing for some at-large voting would, under the circumstances of this case, pass constitutional muster. This is especially true when it is features of the at-large scheme which contributed to the initial finding of dilution. The starting point for such an examination is the observation that while single-member districts are not constitutionally mandated as such, they are definitely the preferred approach. Chapman v. Meier, supra; Connor v. Johnson, 402 U.S. 690, 91 S. Ct. 1760, 29 L. Ed. 2d 268 (1971). This preference may yield, however, where particular circumstances justify a variation. The particular circumstances which would justify such a variation are twofold. The first situation is where "significant interests would be advanced by the use of multi-member districts and the use of single-member districts would jeopardize constitutional requirements. . . ." Zimmer v. McKeithen, 485 F.2d 1297, 1308 (5th Cir. 1973) (en banc). The second situation contemplates use of multi-member (or at-large) districts where they "afford minorities a greater opportunity for participating in the political process. . . ." Zimmer, supra, at 1308. I conclude that the city's plan falls squarely within the second situation suggested by the Fifth Circuit in Zimmer, supra, and permeates to a large extent the first area of circumstances described therein. This conclusion is based upon a consideration of the impact that any plan would have on the Mexican-American citizens of Dallas, intervenors in this cause, and the legitimate governmental interest to be served by having a city-wide viewpoint on the City Council. Chapman v. Meier, 420 U.S. 1, 95 S. Ct. 751, 42 L. Ed. 2d 766 (1975); Fortson v. Dorsey, 379 U.S. 433, 437-438, 85 S. Ct. 498, 13 L. Ed. 2d 401, 404 (1965).
The Mexican-American population of Dallas is approximately 67,073, representing some eight to ten percent of the total. The 1970 census data reflects the eight percent figure. Testimony by plaintiffs' expert and others indicate that Mexican-American census data tends to be understated for a number of reasons, primarily because of the lack of precision in determining who is a Mexican-American. Undisputed is the fact that the Mexican-American citizens are diffused and spread out through all areas of the city. In their case, geographic *793 assimilation has been in large measure achieved. Within the limits of Dallas there are only four census tracts which have a majority Mexican-American population, that is, over fifty percent. The testimony shows that Mexican-Americans have, however, endured some amount of unofficial race discrimination, which has in the past existed in the State of Texas and the City of Dallas. Plaintiffs' expert testified in essence that Mexican-Americans are restricted in their access to the political process but that the restrictions here are to a lesser degree than that for blacks. Witnesses from the Mexican-American segments of the Dallas population, of course, said that the discrimination suffered by them was just as pernicious as any suffered by the black population. The dissenting plaintiff agreed that the Mexican-Americans have suffered some degree of political discrimination in the City of Dallas. The Court finds that Mexican-American citizens of Dallas have suffered some restrictions of access to the political processes within the city but that this restriction does not amount to present dilution. In the case of the Mexican-American citizens, the restrictions involved are difficult to precisely evaluate because of the diffuse residence pattern of the Mexican-Americans throughout the city and the population totals involved. What is not difficult to ascertain is the fact that Mexican-Americans do have a significant interest in the subject matter of this litigation and the Court finds that intervenors are an identifiable minority group with the constitutionally protected right not to be invidiously excluded from participation in the political life of Dallas. White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314, 325-26 (1973). The restriction of access which is present for the Mexican-Americans is of a similar nature to that this Court has found to exist for the black voters of Dallas, but it differs in at least two aspects. In the case of black voters, one factor in establishing dilution was the operation of white dominated slating groups; for Mexican-American voters, any restriction of access caused by slating group operations in the at-large system would not be alleviated by the institution of exclusive single-member district voting, in and of itself. Additionally, the slating group operations, while in some respects restricting access for Mexican-American citizens, also permit them as a group to operate in a "swingvote" manner and give them opportunity they might not otherwise have had. On balance, I conclude that Mexican-American citizens do not suffer from present dilution of their voting strength, and in fact, benefit to a significant extent from at-large voting. Cf., Bradas v. Rapides Parish Police Jury, 508 F.2d 1109 (5th Cir. 1975). This Court is unwilling to penalize the Mexican-American community for its successful efforts to date in assimilating to the degree they have into aspects of the life of Dallas.
In an exclusive single-member district plan the Mexican-Americans must, because of their lesser numbers, form voting coalitions with either black or white voters; that is, because of their diffuse resident pattern, Mexican-Americans will always be a minority group with either a black or white majority. They are in a double-bind situation. The Court finds that an exclusive single-member district plan would do nothing to increase the opportunity for Mexican-American participation in the political life of Dallas and might tend to decrease it. Because of the voting patterns in Dallas, i.e., black voters tending to vote for blacks, white voters voting for whites, the browns will invariably look to coalitions with either blacks *794 or whites in order to maintain political viability. This is a political fact of life for the Mexican-American voters in the City of Dallas. I find that presently Mexican-Americans have to some degree entered into the policy-making structure of the C.C.A. and other slating groups, and have their voices heard on the highest levels of these organizations as well as within the city government. In other words, the Mexican-Americans have entered to some extent into the political life of Dallas through the present at-large system. At-large voting may operate in part as a restriction of access for Mexican-Americans as it has been for blacks. At the same time however, it is clear that at-large voting, offers features which allow greater participation in the political processes within Dallas for Mexican-American voters, that would be unavailable in an exclusive single-member district voting plan.
Significant interest of both black and Mexican-American segments of the voting population will be advanced when the city's plan is implemented. Eight single-member districts will permit black voters a greater degree of access and participation in the political process of Dallas. Likewise, the eight single-member districts will, in combination with the three at-large positions, enhance the opportunity of the Mexican-American citizens of Dallas, to utilize their voting potential in a significant new way, while not undermining the degree of participation they have enjoyed under the exclusive at large voting plan. Mexican-American citizens will, under the eight/three plan, have a heretofore unavailable flexibility and greater opportunity to participate in the political life of Dallas. Zimmer v. McKeithen, supra, 485 F.2d at 1308. Any redistricting plan which did not consider the interest of the Mexican-American community would be suspect from a constitutional point of view. Here we deal with two separate minority groups, each with a significant interest. This interplay of distinct minority interests is markedly similar to the "singular combination of unique factors" which justified a multi-member apportionment plan in Mahan v. Howell, 410 U.S. 315, 333, 93 S. Ct. 979, 35 L. Ed. 2d 320, 335 (1973).
There is another consideration which supports some measure of at-large participation in Dallas municipal government. That is the need for a city-wide view on the part of council. The council has responsibility for policies which affect the city as a whole, as well as those which affect specific geographic and demographic parts. Several members of the present council and the present City Manager presented the view that having some members of the City Council elected on a city-wide basis would be desireable. This desire for some at-large seats was said to be because of the need for a non-sectional viewpoint in resolving matters such as zoning (the Council functions as a zoning appeal board), budgetary considerations (the Dallas City budget is formulated on a city-wide basis) and city planning. Concern was expressed that exclusive sectional or district voices being heard on these matters would be detrimental to the city's interest as a whole. The Court finds that these interests are of significance in considering any remedy which might be advanced. Chapman v. Meier, supra, 420 U.S. at 20, 95 S. Ct. 751. That there is some need for a city-wide interest to be maintained in the government of Dallas is admitted by plaintiffs in their presentation of the ten/one plan. Here, as noted before, one council seat, designated as mayor, is elected city-wide. The evidence is controverted on the point of whether or not the mayor of Dallas is merely a ceremonial mayor with essentially council duties. What is *795 undisputed is that the mayor does have the same council duties as any other council member, and to that extent, he may be considered just a councilman. So it appears that it is merely a question of degree as to how much city-wide representation on these legitimate issues is to be present; three as with the city's plan, or one with the plaintiffs' ten/one plan. The Court believes and so finds that there is a legitimate governmental interest to be served by having some at-large representation on the Dallas City Council; that this governmental interest is the need for a city-wide view on those matters which concern the city as a whole, e. g., zoning, budgets, and city planning; and that three at-large members do not render the city's plan constitutionally infirm.Chapman v. Meier, supra; Fortson v. Dorsey, supra. Furthermore, as the Supreme Court has recognized, at-large representation may be of benefit to all persons of the area from which the particular legislator is elected. "[S]ince . . . [the] . . . tenure [of the legislator] depends on the county-wide [in our case city-wide] electorate, he must be vigilant to serve the interest of all the people in the county [city], and not merely those people in his home district." Fortson v. Dorsey, 379 U.S. 433, 85 S. Ct. 498, 501, 13 L. Ed. 2d 401, 404 (1965). The case sub judice presents just this type of situation wherein at large representation will be of significant benefit because of the particular nature of the city-wide interest to be served.
The demographic pattern of the eight districts in the City's plan is:
Population Deviation Percentage of Council Total from Mean Average of Total Population District Population 105,551 Black Mex.-Am. 1 105,599 +.00045 2.08% 7.63% 2 105,529 -.00020 25.90% 20.00% 3 105,759 +.00197 5.96% 7.00% 4 105,676 +.00118 3.63% 3.20% 5 105,433 -.00112 0.26% 10.60% 6 105,604 +.00050 73.60% 7.20% 7 105,353 -.00196 0.40% 5.00% 8 105,448 -.00098 87.30% 3.44%
*796 There was some initial concern by the Court by the relatively high concentration of black voters in District 6 (73.60%) and District 8 (87.30%). This concern was directed to the question of whether this concentration presented cluster dilution. It is relatively easy to find cluster dilution where it can be shown that district lines are drawn with a purpose or intent to racially discriminate. Wright v. Rockefeller, 376 U.S. 52, 84 S. Ct. 603, 11 L. Ed. 2d 512, 515 (1964); Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960); Howard v. Adams County Board of Supervisors, 453 F.2d 455, 457 (5th Cir. 1972). In this case, however, the evidence is clear that racial gerrymandering was not the purpose or intent of the district lines formulated in 1972 and readopted by the city as a result of this litigation. I find that the eight districts of the city's plan follow natural and rational boundaries and that no gerrymandering is present. Generally, the boundaries of each district are major thoroughfares, rivers, creeks, or city limits. For example, a significant portion of the boundary between Districts 4 and 5 (two "white" districts) is Northwest Highway (Loop 12). Likewise, a good part of the district line between Districts 5 and 7 (again two white districts) is the G.C. & S.F. Railroad right-of-way. The Trinity River is the district line separating Districts 7 and 8 (a "white" and a "black" district, respectively); the river also separates two "black" districts, 6 and 8. Finally, the boundary between Districts 6 and 7 (a "black" and a "white" district) is in large measure, White Rock Creek. The only possible factual situation presented by the evidence which could support a finding of racial intent in drawing district boundaries is the percentages of minority voters contained in each district. In this case I find that although the percentage of black voters contained in Districts 6 and 8 is somewhat high, the concentration in and of itself does not amount to dilution of the black vote. The Supreme Court encountered a similar situation in Wright v. Rockefeller, 376 U.S. 52, 84 S. Ct. 603, 11 L. Ed. 2d 512 (1964). Plaintiffs in that case argued for a finding of racial intent in drawing boundaries and attempted to show that it "was impossible to have districts . . . [as presented there] . . . unless they `were drawn with regard to race'. . . ." Wright v. Rockefeller, supra, at 54, 84 S. Ct. at 604, 11 L. Ed. 2d at 514. There were four districts involved in Wright One was 94.9% white. One was 86.3% black and Puerto Rican and two were approximately 72% white. The Court held that the plaintiffs in that case failed to show that the district lines were a product of a "state contrivance to segregate on the basis of race. . . ." Wright v. Rockefeller, supra, at 58, 84 S. Ct. at 606. There has been much controverted testimony in the present cause over how best to draw district lines, whether with an eight/three, ten/one, or eleven/zero plan. Dallas presented testimony from citizens of the black and Mexican-American community saying that the city's plan was a plan which would allow minority *797 access to the political process in a meaningful manner. Plaintiffs' witnesses sharply disputed this testimony. The dissenting plaintiff in turn offered evidence tending to show that both of plaintiffs' plans fragmented the vote of South Dallas, an area which had many unique and common concerns for the black community. This testimony led the Court to fully comprehend that "there is no agreement on whether the political interest of a minority group are best maximized by an overwhelming majority in a single district, bare majorities in more than one district or a substantial proportion in a number of districts. . . ." Turner v. McKeithen, 490 F.2d 191, 197 n. 24 (5th Cir. 1973). I find that in Dallas, the constitutional as well as the political opportunity of both minority groups affected are best maximized by the city's plan. This Court is not constitutionally required to see that any particular number of "safe" seats are provided for either black, brown or white citizens of Dallas. What is required is an apportionment plan that does not dilute, minimize, or cancel out the voting strength of any group. White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973). Any plan must provide for equal access to the political process for all citizens of Dallas. It is the duty of the city's legislative body to come forward with a plan that will provide this access, and I find that the City of Dallas has met its obligations in coming forth with its eight/three plan.
This Court believes what the Supreme Court has said time and time again in reapportionment cases regarding the responsibility of state legislative bodies to reapportion themselves. "[R]eapportionment is primarily the duty and responsibility of the state through its legislature or other body rather than of a federal court . . . ." Chapman v. Meier, supra, (citing Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506 (1964); Maryland Committee v. Tawes, 377 U.S. 656, 84 S. Ct. 1429, 12 L. Ed. 2d 595 (1964)). I cannot reemphasize this point too strongly. The Supreme Court in Chapman v. Meier, supra, remanded that cause with a high hope that the North Dakota legislature would accept its duty to enact a constitutionally acceptable reapportionment plan. Only then, if the legislature failed in that task, would the responsibility fall to the federal courts.
Testimony shows that the eight/three plan was not a hasty or ill-considered one. The Dallas City Council has had many redistricting plans before it in the last two or three years. Various citizen groups have had opportunities from time to time to voice their opinions on general types of apportionment plans. The members themselves have offered council several redistricting plans during the past two years. The Court finds that the city's plan was merely one of many which has been before the City Council for some time, and not one which was hastily conceived merely for the purposes of this litigation.
The Court is not unmindful of its role in apportionment cases. Absent particularly pressing circumstances justifying at large voting schemes, I would not hesitate to approve only single-member districts. Connor v. Johnson, supra. This would be especially true if there had been no history of at-large voting in Texas and in Dallas. I find, however, that at-large voting, especially on the municipal level has been an integral part of Texas local governments and that at large voting schemes have their genesis in reasons other than those racially motivated. In other words, this Court is not forcing at-large voting on the citizens *798 of Dallas with its acceptance of the eight/three plan because Dallas has had at large voting since at least 1907. There is a preference for single-member districts in considering for constitutional purposes apportionment plans. However, at-large plans have never been declared constitutionally invalid per se. Connor v. Johnson, 402 U.S. 690, 91 S. Ct. 1760, 29 L. Ed. 2d 268 (1971); Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858, 29 L. Ed. 2d 363 (1971). I, too, decline to hold that a plan which provides for some at large representation is in itself constitutionally defective where it was only certain aspects of the city's exclusively at large plan which rendered it constitutionally impermissible. Lucas v. Colorado General Assembly, 377 U.S. 713, 731, 84 S. Ct. 1459, 12 L. Ed. 2d 632, 644 n. 21. Those defects are eliminated by the city's eight/three plan. In Dallas, under the factual situation presented here, neither all single-member districts or exclusive at-large voting offers the balance which is necessary so that all citizens may have equal opportunity of access to the political process. Both plans offer advantages to each minority group as well as to the white majority. The eight/three plan allows the benefits of both schemes without the potential for mischief which is present under each exclusive plan.
At the close of testimony on February 8, 1975, I made findings which approved the city's plan as constitutional. Accordingly, I ordered that the eight/three plan as had been offered by the city, be instituted in time for the April 1975 city council elections, and that the election proceed on the scheduled date. Upon request from the City, the Court modified those provisions of the Dallas City Charter which provide that council candidates have petitions containing signatures from three hundred individuals qualified to vote for the proposed candidate, to allow petitions containing signatures from any three hundred qualified voters residing in the City of Dallas be sufficient for candidacy. This modification is for the April 1975 election only.
On February 21, 1975, plaintiffs filed their motion for attorney's fees and costs. This motion was later supported by appropriate affidavit and on February 27, 1975, attorneys for Mexican-American Intervenors filed their motion on this matter. The City of Dallas filed its opposition to these motions on March 14, 1975. The Court has the matter of attorney's fees under advisement and will make its ruling in a supplement to this opinion.
The Court has previously declared the City of Dallas' exclusively at-large voting scheme for Dallas City Council constitutionally invalid and approved a combination single member district, at large voting plan. The combination plan was initially proposed by defendants as members of the Dallas City Council, after the Court afforded them an opportunity to do so. On March 25, 1975, findings of fact and conclusions of law were entered by the Court. That opinion and order is adopted and incorporated for all purposes in this order concerning attorneys' fees and costs. In the March 25, 1975, opinion I kept under advisement the matter of attorneys' fees and costs. Now, having reviewed the entire record in this matter and considering the briefs of the respective parties I conclude that certain costs should be allowed plaintiffs and that an award of attorneys' fees is inappropriate and will not be allowed.
Plaintiffs and Intervenors moved for an award for attorneys' fees and costs initially contending that they were entitled to such an award under two theories. First, because they were *799 vindicating significant rights for the citizens of Dallas in eliminating an unconstitutional voting scheme, they were acting as private attorneys general and thus entitled to an award. And, second, an award is proper because of the defendants' bad faith in the conduct of this litigation.
The private attorney general theory has of course been recently foreclosed. In Alyeska Pipeline Service Company v. Wilderness Society, 421 U.S. 240, 95 S. Ct. 1612, 44 L. Ed. 2d 141 (1975), the Supreme Court extensively reviewed the conceptual and historical structure for awards of attorneys' fees under the private attorney general theory, and concluded that in the absence of a specific statute, it would be improper for the District Court to award attorneys' fees on this basis. Significantly, the Supreme Court expressly disapproved the authorities sustaining private attorney general awards. These authorities are of course the same ones originally cited by movants in support of their motion. Plaintiffs now strongly advance the alleged bad faith of defendants to support an award of attorneys fees. There is no doubt but that a finding of bad faith or obdurately obstinant conduct on behalf of defendants would support an award of attorneys' fees. Sims v. Amos, 340 F. Supp. 691, 694 (M. D.Ala.1972) (Three Judge Court) aff'd. sub nom. Amos v. Sims, 409 U.S. 942, 93 S. Ct. 290, 34 L. Ed. 2d 215 (1972). In this case however, I find there is simply no factual basis to justify a finding of bad faith.
This litigation has been complex and protracted. Conduct of defendants and their counsel was at all times exemplary and most helpful to the Court as the case developed over its numerous contours. Very few of the matters before the Court were simple or easy to unravel. Many legitimate factual and legal issues were properly raised by defendants for consideration. None can be said to have been frivolous or completely without merit. Much of the factual framework developed at the first part of the hearing on the constitutionality of the old exclusively at large voting plan, assisted the Court in the remedy portion of this litigation. Defendants' contentions that the old plan met constitutional standards did not survive plaintiffs' attack. Many of these same contentions, however, supported the Court's later acceptance of the eight-three, mixed plan which was formulated by the City. Failing to prevail does not amount to bad faith. Compare Doe v. Poelker, 515 F.2d 541 (1975), wherein the Court noted that the history of that litigation revealed a "wanton, callous disregard for the constitutional rights [of plaintiffs therein] . . . ."
The voting system of Dallas is established by the City Charter which was *800 adopted by a majority vote of the voters of the City. Changes to the voting system necessarily are changes to the Charter and absent a judicial determination of unconstitutionality, such changes can only be effected by a Charter Amendment adopted by the voters.
It is a City Charter provision which was at issue here, not a mere ordinance. A majority vote of the registered voters of the City is required to change the Charter. The members of the City Council, cannot, on their own modify the Charter to alter the voting scheme. Council's role is to propose changes to the Charter by proposed amendment which in turn is approved or rejected by the voters of Dallas. In this context, after the initiation of this suit, in 1973, Council did propose a Charter Amendment which would have provided, inter alia, for an enlargement of Council and for a mixed, at large, single member district voting plan. This scheme of enlargement and modification of the voting system was overwhelmingly rejected by both black and white voters. From this factual vantage point, I do not believe that bad faith or obdurately obstinate behavior by defendants has been demonstrated in any way. If defendants had attempted to unilaterally change the voting plan, absent a vote of the people or a ruling from a court with proper jurisdiction that a provision of the Dallas City Charter was constitutionally invalid, they would have been acting unlawfully. Compare, Sims v. Amos, supra, and cases cited therein for examples of bad faith. In Sims, the Court noted that the history of that litigation was "replete with instances of the Legislature's neglect of and even total disregard for, its constitutional obligation to reapportion. . . ." Sims v. Amos, supra, 340 F. Supp. at 693-94. There, as in other reapportionment cases the legislative body had the power and the obligation to reapportion. In the instant case, it is apparent that the Dallas City Council could not have lawfully changed the voting scheme on its own. It lacked the authority to do so. Where there exists a duty and the existing power of an entity to reapportion itself and that entity wilfully and repeatedly refuses to do so, as was evidenced in some legislative redistricting cases, I agree that those circumstances go far to establish bad faith. Where however there is an absence of inherent power to effect such an apportionment, different circumstances are presented. There is absolutely no basis for concluding that defendants have presented their case arbitrarily or capriciously. No racial motivation or intent has been found. In such circumstances, no bad faith exists.
Advancing serious defenses and litigating contested facts is not bad faith. Defendants were in part unsuccessful in litigating complex issues. When the Court announced its finding of unconstitutionality, defendants were given the opportunity to come forward with a constitutional plan. They promptly submitted their mixed, at large, single member district eight/three plan. I have found that this plan meets constitutional standards. Chapman v. Meier, 420 U.S. 1, 95 S. Ct. 751, 42 L. Ed. 2d 766 (1975). I believe that this presentation of a constitutional voting plan by City Council to be one more demonstration of defendants' good faith. Wallace v. House, 515 F.2d 619 (5th Cir. 1975) (concurring opinion).
Plaintiffs, having prevailed in their suit to have the old at large voting scheme declared constitutionally infirm are entitled to their costs as allowed by law. 28 U.S.C. § 1911 et seq. Rule 54(d), Fed.R.Civ.P. The following enumerated costs as shown on plaintiffs' Bill of Cost filed February 21, 1975, will be allowed.
*801 Fees of the Clerk $ 40.00 Fees of the Marshal 10.12 Fees and Disbursements for Printing 1,106.97 Fees for Witnesses 224.90 Fees for Exemplification and Copies of Papers 104.50 Costs Incident to Taking of Depositions 1,683.38 _________ TOTAL $3,169.87
The Court feels that the fees requested for plaintiffs' expert witnesses are not properly taxable. Henkel v. Chicago, St. Paul, M. & O. Ry., 284 U.S. 444, 52 S. Ct. 223, 76 L. Ed. 386 (1932).
It is so ORDERED.NOTES
 Tex.Const. Art. VI, § 3, Vernon's Ann.St., in essence provides that all qualified electors within a city shall have the right to vote for all elective officers of a city. See also, Tex. Const. Art. VI, § 3, and Vernon's Tex.Rev. Civ.Stat.Ann. art. 1175 (1963). See however, State v. McAllister, 88 Tex. 284, 31 S.W. 187, 189 (1895), which holds Tex.Const. Art. VI, § 3 not to be a bar to the election of city aldermen by wards. On July 22, 1974, Chief Judge Brown of the Fifth Circuit advised the Court that he had declined to constitute a three-judge court in this cause.
 United States Census Data, 1970. Unless otherwise stated 1970 census data is used for all population and percentage figures.
 The Court also notes that participation by the black community in the political processes of the city has generally improved in the past few years. In other areas, particularly, responsiveness of Dallas to minority concerns, improvement has been spectacular during this same period. The Court believes however that post litigation changes in these regards are similar to the argument that the electoral success of blacks forecloses dilution attacks. This contention has been specifically rejected as to any bearing on dilution itself. Zimmer v. McKeithen, 485 F.2d 1297, 1307 (5th Cir. 1973) (en banc). In evaluating the new plan which the city has presented, however, this improvement is of significance.
 Uncontradicted testimony shows that CCA is an election year phenomenon and is only concerned with Dallas council elections. During off election years the association has no permanent governing body or officer except the president. With the approach of a council election, the president initiates activity by naming various election committees which in turn hold meetings and select candidates. Historically, CCA grew out of a reform movement in the 1930's which was aimed at correcting abuses of the old commission form of government by infusing citizen participation into governmental affairs.
 The racial composition of the present city council is two blacks (one male and one female); one Mexican-American (male) and eight whites (one of whom is female). The Court notes this composition as being one more indication that circumstances have changed for the better since the institution of this lawsuit and while of not the quantity to overcome the Court's finding of dilution, these changes are again of consequence in considering the proper remedy in this matter. See note 2, supra.
 These factors include a showing that minority residents are frequently ignored in the selection of candidates ; slating group operation combined with a high correlation between endorsement and election, and the high cost of city-wide campaigning. Lipscomb v. Jonsson, 459 F.2d 335, 339 (5th Cir. 1972).
 The trial court in Robinson v. Comm'r Court, Anderson County, supra, apparently found that the dilution present in that case was the result of a purposeful scheme "designed precisely to dilute the black vote . . ." This Court stresses that in the case of the City of Dallas that there has been no evidence of any design or purposeful intent to dilute the vote of any of its minority citizens. Rather, it is in the operation of the exclusive at-large election plan in combination with the existence of a historical pattern of discrimination which supports the finding of dilution.
 See, however, notes 2 and 4, supra.
 The population variance of the eight districts is not in issue here. The district boundaries were redrawn in 1972 partly in response to this litigation. Plaintiffs' original claims included allegations of unproportional districts and population inequality. After the City Council redrew the district lines plaintiffs subsequently abandoned these contentions. The Court finds that there is substantial equality of population among the eight districts presented by the city's plan. Upon its initial presentation to the Court, the city's plan met objection as to the lack of precision in district boundaries. The city subsequently tendered its plan with modifications to meet these objections. See, defendant's exhibit No. 95 showing census tract composition of districts and defendant's exhibit No. 96 showing so-called field note description of boundaries.
 Mexican-American intervenors presented no plan as such, which in their view would alleviate their situation. Generally, they supported the single-member district concept, but testimony clearly showed that in order to have a single-member district plan in Dallas wherein Mexican-American citizens constitute a majority in a single-member district at least twenty districts would be required. Other suggestions included some form of cumulative voting for council seats, concentric districts around the city and non-contiguous council districts. The Court believes that these suggestions are completely void of merit and would do nothing constructive for either the political interest of minorities or the constitutional interests of all citizens of Dallas.
 These are technical matters of census classification. For example, in some instances being a Spanish surnamed citizen is a basic standard; other considerations may be whether or not the person speaks Spanish in the home or whether Spanish is the native language of the person. Apparently, the totals used by the city even include a very small number of citizens who are of Puerto Rican origin.
 Compare this assimilation with the high concentration of black citizens residing in the inner city area. Testimony suggests that single-member district voting tends, in some cases, to amplify segregated housing patterns. It would be indeed ironic if a remedy for dilution of voting strength resulted in a major impediment to minority citizens enjoying their constitutional rights. See, page 5, supra.
 Compare Chapman v. Meier, supra, 420 U.S. at 19, 95 S. Ct. 751 where there was no question of any apportionment plan affecting different minority groups differently.
 All factions in this lawsuit provided the Court with testimony from black, white and Mexican-American citizens in support of their particular plan (s).
 Budget and services certainly do not stop at district boundaries. Other areas of particular concern for Dallas as a whole include downtown revitalization projects, convention business and airport services.
 The Court has found dilution of the black vote. This is not to say however, that black citizens have not, to a certain degree, made the at-large system work for them as have the Mexican-Americans. Within the framework heretofore described, black citizens of Dallas have entered into the political process. What the Court is attempting to avoid is approving any remedy which does not, on balance, enhance constitutional opportunities for all citizens of Dallas. The Court is particularly concerned with the prospect of district sectionalism which usually occurs in an exclusive single-member district plan. The Court is convinced that no matter how many single-member districts are drawn in Dallas, black voters in all probability would never elect more than 25% of city council so long as the present pattern of voting exists. With all single-member districts and the present voting pattern, it would be possible for a majority of council to "freeze out" this 25% and for all practical purposes ignore minority interests. Some at-large seats will give all minority voters the capacity to function as a swing vote and have their voice heard. The present responsiveness of city council demonstrates that a politically aware council does not choose to ignore significant groups of voters when its electorate is the entire city. Fortson v. Dorsey, supra.
 Compare with fragmentation dilution as presented in Robinson v. Comm'nr Court, Anderson County, 505 F.2d 674 (5th Cir. 1974).
 See, note 7, supra.
 See, note 9, supra.
 It is apparent that different district lines could be drawn under an eight/three plan to achieve different proportions of minority population. In fact, city council considered and rejected an alternative eight/three plan which gave black voters a majority in three districts (one with approximately 66%; one with 62% and one with 61%). There was some testimony on what constituted a "safe" district, but this Court is not concerned with safe districts, only constitutional ones. Whitcomb v. Chavis, 403 U.S. 124, 156-60, 91 S. Ct. 1858, 29 L. Ed. 2d 363, 383-85 (1971); Turner v. McKeithen, 490 F.2d 191, 197 n. 24 (5th Cir. 1973). This Court sees no impediment to council drawing different district boundaries after the next scheduled election and in fact, there may well be under Chapter 4, Section 5 of the Dallas City Charter, an obligation of council to periodically review district boundaries.
 Plaintiffs' expert is white and was severely criticized by the members of plaintiffs' class who opposed the ten/one and eleven/zero plans. The implication of the criticism was that plaintiffs' plans had concerns other than those of the black community or constitutionality in formulating district lines. See, Robinson v. Comm'nr Court, Anderson County, 505 F.2d 674 (5th Cir. 1974), for an example of fragmentation of the black vote. The Court finds that the city's plan best preserves the integrity of the area known as South Dallas, which has an identifiable set of common interests.
 Compare, Chapman v. Meier, supra, where the District Court formulated its own plan for the North Dakota legislature and that state had no tradition for multi-member districting.
 Intervenors, representing the Mexican American citizens of Dallas participated only in the remedy phase of this cause after the old exclusively at large voting plan had been declared invalid. See, this Court's Order of March 25, 1975.
 Brandenburger v. Thompson, 494 F.2d 885, 890-93 (9th Cir. 1974) (concurring opinion) is also of interest. Here, it was observed that the now disapproved of cases, which sustained an attorney's fee award predicated on a private attorney general concept, also had strong elements of bad faith present. In other words, these cases were saying although we don't need to find bad faith to award attorneys fees, we do find it present and thus will justify an award of attorneys fees on both concepts.
 This is not to detract from the conduct of movants in this suit. It is, however, defendants' bona fides which is now at issue.
 Council member defendants urged, as did Mexican-American intervenors in the remedy hearing, that any plan consider the needs of each distinct minority in the City. I feel that this attitude alone, negates any finding of bad faith on the part of defendants and tends to show the merits of the City's claim of responsiveness. See also, this Court's Order of March 25, 1975, at pages 16 and 22-23. This Court's finding of dilution was predicated largely upon the realities of housing and voting patterns existing within the City when viewed in an historical context. It is difficult to deny the past, and the defendants, I believe, recognized the past and defended on the basis of present responsiveness. While they did not prevail, I cannot say that the defensive positions were insignificant or advanced for dilatory purposes. As I have herein stated, I do not believe that merely losing on the merits necessarily implies bad faith.
 Cf. Pierson v. Ray, 386 U.S. 547, 87 S. Ct. 1213, 18 L. Ed. 2d 288 (1967). Good faith reliance on a statute can of course be a valid defense to a suit under the Civil Rights Act under 42 U.S.C. § 1983. See, Traylor v. City of Amarillo, 492 F.2d 1156, n. 2 (5th Cir. 1974).
 United States v. Kolesar, 313 F.2d 835 (5th Cir. 1963); See generally, 10 C. Wright & A. Miller, Federal Practice and Procedure, § 2676 (1973).
 Plaintiffs have urged over $13,000.00 as expert witness fees. An amount for expert witnesses, above subsistence and travel is not viewed as a taxable cost. Baum v. United State, 432 F.2d 85 (5th Cir. 1970). In Sims v. Amos, 340 F. Supp. 691, 695 (M.D. Ala.1972) (three-judge court) costs for an expert witness were allowed without consideration of defendant's bad faith on the point because the Court had earlier found bad faith in connection with attorney's fees. Additionally, the Court found that the expert performed a function which otherwise would have had to have been accomplished by a court appointed master. Sims v. Amos, supra, n. 11. In the instant case, plaintiff's principal expert obviously had considerations other than those of constitutionality, or those of the black community in mind when he formulated plaintiffs' ten/one or eleven/zero plan. There was testimony indicating that these plans tended to fragment the black vote of South Dallas. See, this Court's Opinion of March 25, 1975, n. 21. In any event, considering the lack of bad faith of defendants and the reasoning of Alyeska Pipeline Service Co. v. Wilderness Society, supra, I feel expert witness fees are inappropriate and will not be allowed. See generally, 6 J. Moore, Federal Practice, ¶ 54.77 [5.-3] (2nd Ed. 1974).