This case involves a racial gerrymander not by a state legislature but by a federal district court. The trial judge was well intentioned, of course, but his plan for drawing the boundaries of four state senate voting districts in New Orleans cannot be considered “benign districting” as the Supreme Court used that term in remanding this case to the Court of Appeals.

This litigation started as a frontal attack on the self-reapportionment of the Louisiana legislature under Acts 107 and 108 of 1970. In 1971, in five suits the plaintiffs attacked the legislative reapportionment plan as violative of the one man, one vote principle and as racially discriminatory. The trial judge, the Honorable E. Gordon West, consolidated the cases and appointed a special master, Edward J. Steimel, to recommend a plan that would comply with legal standards for legislative apportionment.1 The trial judge approved the Steimel plan without any deviation. 333 F.Supp. 452. This Court affirmed the judgment, except with respect to four senatorial districts in Orleans and three in East Baton Rouge Parishes.2 *895As to these seven senatorial districts, we approved alternate plans submitted by interested state senators. The judgments of both courts represented an almost total defeat for the Louisiana Attorney General, who had defended the statutory reapportionment and had attacked the Steimel plan largely on the ground that it contained all single-member districts in both Houses, in disregard of state policy allowing multi-member districts.

The litigation has now narrowed to a dispute over senate districts 2, 3, 4, and 5, four of the seven senate districts in Orleans Parish. (The parish is coterminous with the City of New Orleans.) This Court approved a plan for these districts proposed by incumbent State Senators Adrian G. Duplantier, Ignatz G. Kiefer, Michael H. O’Keefe, and Theodore H. Hickey. The first three of these senators, all attorneys, were associated by the Attorney General as co-counsel for the State after we had remanded the case to the district court for a hearing on the special master’s plan; the district court first approved the plan without a hearing. The real adversaries here are these senators 3 and Mrs. Dorothy Taylor, a black member of the Louisiana House of Representatives, who brought one of the class actions attacking Acts 107 and 108.

The United States Supreme Court, on Mrs. Taylor’s petition for writ of certiorari, remanded the case to our Court, 407 U.S. 191, 92 S.Ct, 1980, 32 L.Ed.2d 648, "because this record does not fully inform [the Supreme Court] of the precise nature of the litigation and because [the Supreme Court has] not had the benefit of the insight of the ‘Court of Appeals’.” The Supreme Court stated that the Court of Appeals “without opinion . . . adopted the Attorney General’s alternative division of New Orleans”.4 The Supreme Court, in its per curiam opinion, said:

An examination of the record in this case suggests that the Court of Appeals may have believed that benign districting by federal judges is itself unconstitutional gerrymandering even where (a) it is employed to overcome the residual effects of past state dilution of Negro voting strength and (b) the only alternative is to leave intact the traditional “safe” white districts. If that were in fact the reasoning of the lower court, then this petition would present an important federal question of the extent to which the broad equitable powers of a federal court, Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 15 [91 S.Ct. 1267, 28 L.Ed.2d 554], are limited by the colorblind concept of Gomillion v. Lightfoot, 364 U.S. 339 [81 S.Ct. 125, 5 L.Ed.2d 110], and Wright v. Rockefeller, 376 U.S. 52, 57, 67 [84 S.Ct. 603, 11 L.Ed.2d 512].

In this Court’s short, earlier decision we did not reach the important federal question that concerned the majority of the Supreme Court.

*896The district court, in effect, admittedly approved Steimel’s structuring of the four districts for the purpose of maximizing the black voting strength in two districts to ensure two safe black senate seats. But this racial gerrymander rested on a premise that was patently erroneous. It is true that the Senators’ plan relies heavily on historical boundaries, wards in the City of New Orleans. It is also true that the trial judge said: “as brought out in the evidence, it was this policy and these ‘historical’ boundaries that produced but two negro legislators during the Twentieth Century”. He concluded: “ ‘[Historical’ boundaries of voting districts i.n Louisiana reflect[ed] a history of racial discrimination. Adherence to the historical boundaries alluded to by objectors [had] been the prime reason why only two negroes [had] been allowed to sit in the Louisiana Legislature in the last 75 years.” The Supreme Court quoted this language in its opinion. It is understandable that this language should (unintentionally, of course) mislead the Supreme Court.

There is absolutely nothing in the record to support such a “finding”. As a statement of Louisiana history, it is an error of monumental magnitude.

The heavy concentration of black votes in Steimel districts 2 and 4 could be accomplished only by diluting the black vote to negligible effectiveness in districts 3 and 5 which would then become super-safe white districts. The Steimel districts disregarded historical boundaries and were as multi-sided as the “uncouth” gerrymandered district in Gomillion v. Lightfoot, 1960, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110. The Senators’ plan observed historical ward boundaries as nearly as possible, but differs as radically from the statutory plan embodied in Act 108 as jt differs from the Steimel plan.5 In contrast to the latter, the districts are compact, contiguous, mini-sided; and, considering the shrinking white population, the increasing black population, and the accelerating black registration in New Orleans, the Senators’ plan gave black voters in the four districts better access to participation in the election of state legislators than the Steimel plan. We did not and do not think that “the only alternative [to the Steimel plan was] to leave intact the traditional ‘safe’ white districts”.

Historically, there has never been any nexus whatever in Louisiana between the establishment of traditional political boundaries and the denial of access of blacks to the state legislature.6 In this century, until this Court compelled parish registrars of voters to register blacks and until the Voting Rights Act of 1965 was enacted and enforced, blacks could not be elected to the Louisiana legislature — to be blunt — because there were no black voters. It is as simple as that. Since adoption of the Louisiana Constitution of 1898 and until recently, the legislature disfranchised blacks overtly; it was never necessary for the legislature to resort to covert disenfranchisement of blacks by manipulating the boundaries of legislative voting districts.

This case, therefore, was not one of benign districting to overcome the residual effects of discriminatory districting. And the Senators’ alternative plan was, in our opinion, more effective than the district court’s gerrymander in allowing blacks access to the political processes involved in electing lawmakers responsive to the needs of their constituents-. Instead bf the case presenting the important constitutional and federal questions the Supreme Court raised, the case presented the question whether the district court’s purposeful, racial gerry*897mander was an abuse of judicial discretion.

On remand, this Court requested and received additional briefs. The New Orleans Senators, still in their capacity as co-counsel for the State, and Mrs. Taylor were, of course, the only parties to file briefs on the remand. We have fully reconsidered the case. With due deference to the Supreme Court, we must say that the record and Louisiana history compel us to adhere to our earlier decision. We see no purpose to be served by remanding the case for the district court to construct a “less drastic alternative”; for one black has already been elected from the Senators’ district 4, one of the districts the trial judge characterized as a “safe” white district from which it would be “impossible” to elect a black. The population and registration figures suggest that a black senator may also be elected from district 2; and in districts 2, 3, and 5 blacks have sufficient population and registration to have the balance of power.

A month after the Supreme Court remanded the case to this Court, Act 457 of 1972 became law. In this statute the Louisiana Legislature reapportioned itself in accordance with the district court’s plan as modified by the Court of Appeals.

We proceed now from this summary to a more thorough discussion of the case.

I.

As a result of Bannister v. Davis, E.D.La.1966, 263 F.Supp. 202, a three-judge court decision, the Louisiana legislature adopted Acts 3 and 4 of 1966, which reapportioned the Senate and the House of Representatives. In Bannister, the Court warned that multi-member districts “should be avoided whenever possible” because they “tend” toward “diluting voting strength” of minorities. 263 F.Supp. at 209.7 In 1970 the legislature enacted Acts 107 and 108, effective in 1971, reapportioning itself based on the 1970 census. The plan contained many districts with wide deviations from the norm and, notwithstanding the cautions in Bannister, had multi-member districts. Mr. Victor Bussie, a well-known labor leader in Louisiana, filed the first action attacking the reapportionment statutes; he asked the district court to fashion a reapportionment plan that would conform with the one ■ man, one vote principle. Shortly thereafter, Mrs. Dorothy Taylor, with other black plaintiffs, challenged the proposed reapportionment. Two other suits were filed. The three cases were consolidated with Bussie.

The court stayed all proceedings pending submission of the plan to the Attorney General of the United States as required by the Voting Rights Act of 1965, 42 U.S.C. § 1973c. The Attorney General concluded that the plan was racially discriminatory.8 The trial judge had previously appointed Mr. Edward J. Steimel as a special master to prepare a reapportionment plan. Mr. Steimel, Executive Director of the Public Affairs Research Council, a non-partisan organization, is highly respected generally and particularly as an expert in the field of reapportionment.9 Judge West instructed the master to hold hearings and, in devising a reapportionment proposal, to maintain the integrity of political divi*898sion, natural and historical boundaries, “as nearly as possible.”10

The special master held four days of hearings during which over a hundred persons were heard. Later, there were objections that he was biased for and against certain incumbents, and would not discuss alternate plans. The Attorney General and Mr. Bussie moved that he be dismissed as a master because of his bias. There is no evidence of bias, and the court dismissed the motion. The master submitted his plan to the district court on August 24, 1971. The court approved the plan that same day —without a hearing.

The decision was a complete victory for the plaintiffs. The one man, one vote principle was vindicated with remarkably small variations from the norm. What is more, there were no multi-member districts in the Senate or the House. In these major respects the plan was beyond reproach. The plaintiffs promptly moved for a new trial. In denying the motion, Judge West pointed out: “We have here the rather unusual situation of the plaintiffs having been given virtually everything they asked for in these cases, and now they, rather than the defendants, against whom judgment was rendered, are seeking a new trial to set their victory aside.”

This Court, on September 4, 1971, felt compelled to vacate the judgment and remand the case for a hearing before the district court under Rule 53(e)(2) to enable objectors to make known to the court their specific objections to the Steimel recommendation and to propose alternate plans. Time was of the essence. Candidates in the primary elections had to qualify by September 15, 1971.

The district court held a hearing on September 8, 1971. The Attorney General defended Acts 107 and 108, but by this time he had joined hands with Bussie. Bussie submitted no alternate proposals. His counsel and the Attorney General argued against single-member districts, questioned their constitutionality, and contended that the state’s policy of preserving historical boundaries should prevail, even if it resulted in multi-member districts. As noted earlier, the Attorney General, perhaps as a courtesy to Louisiana Senators,11 associated Senators Duplantier, Kiefer, and O’Keefe, all from New Orleans, as co-counsel for the State; Senator Hickey, not a lawyer, testified for the senators’ plan. By no coincidence, these senators are incumbents whose districts would be redistricted; two would oppose each other in Steimel District 3 and two oppose each other in Steimel District 5. They agreed with the Steimel plan as to the new districts 6, 7, and 8 in Orleans Parish but disagreed as to the four other proposed senatorial districts in districts 2, 3, 4, and 5. The Senators suggest that Steimel’s purpose was to get rid of two of the four incumbents; the trial judge thought the purpose of the senators’ plan was to avoid one incumbent running against another. We are not so naive as to doubt that the impact of the reapportionment on their political future affected their motives. But the validity of their alternate proposal stands on its own merits.

Meanwhile the Attorney General had moved that the case be heard by a three-judge court. The district judge denied the motion on the ground that the reapportionment legislation was patently unconstitutional.12

In the hearing on September 8 the trial judge gave everyone an opportunity to be heard. The New Orleans Senators submitted maps showing their alternate plan for districts 2, 3, 4, and 5, and testified as to the merits of their plan. The remand had activated interests in other areas than New Orleans. Senator *899J. D. DeBlieux, for example, proposed an alternate plan for districts, 14, 15, and 16 in East Baton Rouge. Judge West noted that in “all fairness to the De-Blieux plan, it. must be said that it is equally as good, . . . but not necessarily any better” than the Steimel plan. The Republican State Central Committee filed an action opposing the Steimel plan and proposing a plan of its own; the action was consolidated with Bussie and the other cases. Representatives of Jefferson Parish objected to sharing four of its allotted representatives with other parishes. Mrs. Taylor recommended acceptance of the Steimel plan, but, filed two exhibits which, as Judge West said, “merely show that plans could be presented which are within acceptable deviation limits while at the same time giving the negro voters an even greater voter strength than does the court-approved plan”. Mr. Steimel was present, but was not called to testify.13 The trial judge reaffirmed his decision of August 24. He extended to September 16 the filing date for candidates to qualify in primary elections.

This Court heard the expedited appeal on September 14, 1971. The Attorney General and Bussie filed a joint brief on appeal, reiterating their arguments before the district court. The New Orleans Senators and others filed briefs in support of their alternate plans. We issued our judgment on the same day we heard the argument, affirming the district court almost completely. We agreed that a three-judge court was unnecessary. We agreed that the Steimel plan complied with the one man, one vote principle; both the Steimel plan for the entire state and the Senators’ plan for Senate districts 2, 3, 4, and 5 allow deviations from the norm well within acceptable limits. We reaffirmed approval of the principle, for Louisiana, of single-member districts for both houses.14 We disagreed with the district court only with respect to four senatorial districts in Orleans and three in East Baton Rouge Parishes.15

II.

The trial judge was “surprised” when the plaintiffs moved for a new trial. We were astonished when Mrs. Taylor filed a petition for certiorari. In the four senatorial districts here involved, blacks have a greater opportunity to make their voting power felt under the Senators’ plan than under. the Steimel plan. The Senators were more realistic, perhaps more aware of the political aspects of reapportionment than Mr. Steimel and the trial judge.

*900At the start, we note that while New Orleans has been losing total population, the black population of the city has been increasing. Between 1960 and 1970 the total population of New Orleans declined from 627,525 to 593,471, a decrease of 5 percent. In that period the black population grew from 234,931 to 267,308, an increase from 37.2 percent to 45 percent. Statistical Abstract of the United States, Dept, of Commerce, Bureau of the Census (1973), No. 23, p. 23. This is an important fact which the Senators’ plan takes into account. Senator O’Keefe, who represented a two-member district with Senator Hickey, testified:

Q What has been the trend, both population and voter registration-wise?

A The trend has been that in the downtown area, the black population and voter registration have increased notably and show a definite upward trend.

It also shows that the movement of the black vote and the black population from the river to the lake was in substantial numbers.16

He also testified as to the effect of the Steimel plan:

A Well, I can best do that by starting in saying that almost an identical plan to this [Steimel plan] was submitted about five years ago for that particular area by the Citizens’ Councils 17 of the city of New Orleans whose stated purpose was to affect the black vote in that area and to be prejudicial to the black, and it had as its purpose to isolate the lake front precincts, being Lakeview, which has one of the highest concentrations of Citizen Council members at that time —Citizen Council members in the city of New Orleans. So I would say the effect of the Master’s plan, as it deals with ward three, four, five, six, and seven, is to cut across a district which would be sixty per cent black and by four years from now will probably be eighty per cent black and to cut across another district which right now is approximately forty per cent black and within another, oh, I would guess four to eight years, would be at least fifty-fifty and probably sixty-forty. So that the net effect is that there would be a protected white area that would never never have influence of blacks and would be cut in such a way that the black voice would not be noticeably visible in this area, and, in effect, isolate the blacks in the front river precincts and also cut the growing area of blacks, such as lawyers, and doctors, who are moving out to the lakefront, [and around Dillard University]— would be substantially cut from the poorer blacks that live closer to the river. So it would.be separating what may be the leadership group of the blacks from the blacks numerical areas.

Black registration in New Orleans lags behind its potential; in 1971 it was 35 percent of the total registration. This may be accounted for, in part, by the fact that “people in the lower socioeconomic status tend to register in less percentages than do persons of a higher socio-economic status” (testimony of David R. Poynter, Senior Research Asso*901ciate for the Louisiana Legislature Council) and by the disproportionately larger number of blacks below voting age (U.S. Census, 1970). In any event the increase in the black population in New Orleans naturally increases the registration. There is less bloc voting among whites than among blacks. Senator Hickey, for example, pointed out that the President of the Orleans Parish School Board and two black judges were elected in parish-wide elections over white opponents.

All of this adds up to the fact that there is a tipping point short of 50 percent. The strong motivation of blacks to vote and to vote for a black and the less than solid bloc voting of whites, many of whom favor increased black representation in the legislature, could mean that a senate district having 43-45 percent black population in 1971 could elect a black senator in 1975, the next election year.

What the Senators were able to come up with in 1971 was a plan for three districts having black populations close to the tipping point and a fourth district having a black population of 54.4 percent. Integration increases as the years pass. What Steimel proposes is two segregated black districts and two segregated white districts. It is ironical that, as Senator O’Keefe observed, the Steimel plan is identical with an earlier Citizens Councils’ plan.18

In the Steimel and Senators’ plans the percentage of the black population, according to the 1970 census, was as follows:

District Steimel plan Senators' plan

2 64.0 42.6

3 16.0 43.7

4 70.2 54.4

5 21.7 42.0

In 1971 the percentage of black registered voters was as follows:

District Steimel plan Senators' plan

2 51.0 37.6

3 18.0 25.7

4 58.0 44.3

5 20.0 24.0

The trial judge stated that “while the Senators’ . plan probably would meet the one man, one vote standard . . . their plan practically eliminates the possibility of a negro being elected from any of the four districts while the court approved plan at least gives them a fair chance in two out of the four districts”. (Emphasis added.) The Court’s attention has been called to the fact, (indeed, it is public knowledge) that Sidney J. Bartholemy, a black, recently received a clear majority (52 percent) over three white candidates in the Democratic primary in Senate district 4 for the seat vacated by Senator Duplantier. In the general election, April 30, 1974, he defeated his white Republican opponent by a vote of 9,106 to 6,743. By 1974 black registration in district 4 had jumped from 44.3 percent to slightly less than 50 percent (17,258 black voters; 17,959 white voters). In Zimmer v. McKeithen, 5 Cir. 1973, 485 F.2d 1297 (en banc) the majority opinion declared that “an appellate court cannot take cognizance of matters not passed upon by the trial court”, and declared to consider the effect of an election of three blacks under a challenged at-large plan, because the election was held after the district court had approved the plan. We think, however, that it is proper for us to consider facts which are of public knowledge in deciding whether to remand the case for the district court to construct another plan. Moreover in Zimmer, the district court simply applied a per se rule that since the blacks were a majority in East Carroll Parish, an at-large plan could n<\t possibly submerge their vote. The ma*902jority reversed, noting that “although population is the proper measure of equality in apportionment. ... access to the political process and not population was the barometer of dilution of minority voting strength”; 485 F.2d at 1303. Here the trial judge conceded the racial gerrymander but justified it on the mistaken premise that lack of access to the political process was a product of discriminatory districting and on the mistaken conclusion that the Senators’ plan “eliminates the possibility of a negro being elected from any of the four districts”. The senators argued in the district court and argued on appeal that their plan gave blacks more and broader access than the Steimel plan. They laid a predicate in the district court for what later happened; this was not the situation in Zimmer. We do not say that the election of one black senator, the first to be elected since 1898, is necessarily conclusive, for there could be voting districts where the white power structure might successfully support a black candidate for various political reasons.19 Nevertheless, Senator Bartholemy’s election shows that a black had the “possibility” of being elected under the Senators’ plan. Perhaps equally significant is the increase in black registration in only three years, a very important factor which the senators considered, but which apparently was ignored in the Steimel plan.

It was a fair inference in 1971 and is evident now that under the Senators’ plan, districts 2 and 4 are especially unsafe for any white candidate, and, given a few years, for black registration to catch up with potential black voting strength, none of the four districts can be considered a “safe” white district. Even in a district with a white senator there are and will be enough blacks so that they will have a voice in the political processes. On the other hand, while the Steimel plan gerrymanders and guarantees the election of a black senator from each of districts 2 and 4, this can be accomplished only by diluting black voting power in districts 3 and 5 to the point where the white senators from those districts could ignore with impunity the special needs of blacks in those districts.

The two plans cover the same total area for the disputed districts. The maps in the record, copies of which are appended to this opinion, graphically show the superiority of the Senators’ plan. The Senators’ plan adheres to closely established ward lines which have served as the basis for legislative districts for a century.20 The districts are compact, contiguous, and straight-lined districts, without diluting *903the increasing strength of black voting power. The Steimel plan egregiously gerrymanders senate districts 2, 3, 4, and 5, virtually ignoring ward lines.21 Under that plan, district 4 is a 24-27 sided figure (there is a dispute as to the number) that is made up of Wards 1 and 2 and selected densely populated black precincts from five other wards (3, 4, 5, 6, and 7). District 5 has 21-24 sides with precincts from four wards 3, 4, 5, and 7. Portions of Ward 7 are spread through three districts. Ward 8 is segmented, also destroying the integrity of its ward lines. The Ninth Ward has a population of 137,000. This is more than 40,000 persons necessary to constitute an ideal single-member senatorial district comprising 93,415 people. The Steimel plan divides the Ninth Ward, with one section of the Ninth Ward forming a senatorial district with *904small portions of Wards 7 and 8 and another senatorial district with a small portion of Ward 8.

The Senators’ plan, as any districting plan must, divides the Ninth Ward. But District 2 contains all of the precincts in Ward 9 except precincts 9 through 29. District 3 consists of the first eight precincts in Ward 9 and all of Ward 8. The established boundaries of Wards 8 and 9 have existed in Louisiana law for many decades. Of paramount importance, all precincts of Ward 9, which are a part of Senatorial District 3, are immediately adjacent and contiguous with those of Ward 8, and precincts 9 through 29 of Ward 9 are exactly within the census track boundaries of the United States Bureau of Census. District 4 contains Wards 6 and 7. The state policy of including Wards 6 and 7 as a single-member senate district (District 4) has been established since 1921. District 5 contains Wards 1, 2, 3, 4, and 5.

III.

A ward in New Orleans traditionally means as much to its residents as a par-? ish or county does to its residents. The City has been divided into wards since 1805, and most of the ward boundaries are far more ancient than any question of Negro voting strength.22 Canal Street and Elysian Fields were ward boundaries in 1805,23 and are now.24 Almonaster, Esplanade, St. Philip, and Canal Streets were ward boundaries in 1836,25 and are now.

The direct ancestor of the present ward structure was adopted in 1852.26 Ward boundaries have been changed since then only by the addition of new wards to accommodate areas newly incorporated into the city, except for a minor change in 1878 to correct an anomaly and a major change in 1880 when a substantial area was taken from the sixth ward and added to the fourth and fifth.27 That change of 1880 was the last change in the ward boundaries to date. The Home Rule Charter of the City for 1954 has the same ward boundaries as its predecessor, the charter of 1912. It is ludicrous to suggest that the wards were gerrymandered in 1912 to prevent blacks from having access to the legislature. At that time, it is just possible that there were as many as 200 blacks registered in New Orleans,28 but they were ineligible to vote in the decisive white primaries.

The first function of the wards was to serve as the districts from which were elected the aldermen who formed the governing council of the City.29 Since then, they have been used as the basic units of apportionment for representatives in the United States Congress, for presidential electors, for state senators and representatives, for judges and lesser officials of the city courts, for city councilmen, for tax assessors, and for the members of the numerous central or regional committees which form the statutory structure of the political parties.30 The wards have structured the working levels of political organizations. Parties and factions have generally been organized along ward lines31 with ward leaders as major political powers.32

Moreover, the wards are real and important parts of the city’s life and culture. Residents of the City are likely to speak of themselves as living in the *905Twelfth ward, or the Seventh, or the Fourteenth, say, in contexts quite apart from politics; indeed, in the same way that one would say that he lived in Marigny or in the Irish Channel or the lower Garden District.

In 1876 the City was first divided into senatorial districts with ward boundaries.33 The wards continued to be units of which senatorial districts were formed through the reapportionments of 1878, 1879, 1902, 1921, 1966, and 1971.34 There was no reapportionment of the legislature between 1921 and 1966.

The long history and continuing vitality of the wards entitle them to the respect accorded political subdivisions by the Supreme Court in Reynolds v. Simms, 377 U.S. 533, 578, 84 S.Ct. 1362, 1390, 12 L.Ed.2d 506: “A State may legitimately desire to maintain the integrity of various political subdivisions, insofar as possible . . . ”

Bannister v. Davis, E.D.La.1966, 263 F.Supp. 202, 206, accorded them this respect: “A reapportionment plan may and should follow the boundaries of existing wards and parishes as long as it represents a good faith attempt to keep the number of persons as close as possible to the ideal.” In this case, Judge West gave this same instruction to the special master.35

The importance attached by the Supreme Coui’t to partitioning by existing subdivisions is illustrated by Abate v. Mundt, 1971, 403 U.S. 182, 91 S.Ct. 1904, 29 L.Ed.2d 399, in which the Court approved substantial population deviations because they were necessary to accommodate that principle.

IV.

With due deference to the Supreme Court, we consider that Whitcomb v. Chavis, 1971, 403 U.S. 124, 91 S.Ct. 1858, 29 L.Ed.2d 363, should control the case now before this Court.

Whitcomb dealt with reapportionment of the Indiana legislature. In Indianapolis, the district court was able to define an area within which lived enough people to qualify for one senate seat and two house seats. A majority of those people were black, and they shared certain other demographic characteristics giving them as the district court said, a community of interest and entitling them to have the apportionment plan drawn in such a way as to realize their block-voting potential. Accordingly the district court’s plan assigned to that ghetto area two single-member house districts and one single-member senate district. In order to accomplish that result, however, it was necessary for the district court to abandon the longstanding Indiana policy of apportioning the counties as single-member districts.

The Supreme Court considered directly the issues presented by this judicial attempt to enhance the representation of a particular ethnic-socio-economic group. It disapproved. The people of the ghetto were entitled to participation in the political process and to representation as people, and the record did not show that they had been denied that. But they were not entitled to recognition and representation as poor black ghetto dwellers, nor on account of any other characteristics that might be typical of their area. The constitution extends equal protection of the laws to people, not to interests. The Court recognized that disregard of the multi-member districting policy would be justified were it shown to be discriminatory against constitutionally protected rights. But there was nothing in that record to indicate an invidious design in the Indiana practice of apportionment by multi-member districts, as there is nothing invidious in the Louisiana practice of apportioning *906with respect to the wards of New Orleans.

There was, in the record, a showing that few ghetto residents had been elected to the legislature, as there was in this case a showing that, until recently only two blacks had been elected to the Louisiana Legislature in this century. There was a showing in that case which has no counterpart in this case, that candidates favored by the ghetto residents had lost more elections than they had won in recent years. But, the Supreme Court said, there is no constitutional right to be included within a district that is especially favorable to the interests of one’s own group, or even to be excluded from a district that is “safe” for some other group.

The Supreme Court, in its per curiam opinion in the instant case recognized that Whitcomb was similar in some respects, “and, in each case the equitable remedy of the court conflicted with a state policy”. “There the state policy favored multi-member districts whereas here the policy favors maintenance of traditional boundaries.” But, said the Court:

“The important difference, however, is that in Whitcomb it was conceded that the State’s preference for multimember districts was not rooted in racial discrimination, 403 U.S., at 149, 91 S.Ct., at 1872. Here, however, there has been no such concession and, indeed, the District Court found a long ‘history’ of bias and franchise dilution in the State’s traditional drawing of district lines.” Cf. id., at 155, 91 S.Ct., at 1875.

This brings us to the spectacularly false premise on which the district court based its approval of gerrymandering senate districts 2, 3, 4, and 5. Any student of history and certainly anyone familiar with Louisiana politics, such as Senator Hickey, the unwitting source of the district court’s revision of Louisiana history, knows that the district court’s “finding” was not just “clearly erroneous”; it was dead wrong.

V.

There is absolutely nothing of substance in the record and less than nothing in Louisiana history to support the notion that the State’s policy of preserving traditional political boundaries for legislative districts was, in the language of the Supreme Court’s per curiam opinion, “rooted in racial discrimination”.

From 1868 until the federal troops were withdrawn in 1877,36 there were more Negro voters in the state than there were white voters, and the legislature was dominated by the Radical Republicans, whose support came primarily from Negro voters.37 Presumably if any race were to be disadvantaged by districting practices in this period, it would not be the Negro race. Reconstruction ended when the federal troops left, but Negro suffrage did not. For the next twenty-one years, the votes of Negro citizens were earnestly solicited by rival factions.38 After 1868 then, and until sometime in the past decade, Negro voting was simply irrelevant to political power struggles, and to reapportionment. For three decades, from 1868 until 1898, Negro suffrage was certainly not irrelevant. It was perhaps the most important fact of political life, but it was never dealt with, to enhance or to diminish its power, by reapportionment. One may search the voluminous political literature of this era in vain for any reference to reapportionment in relation to race.

This Court has studied the history of the enslavement, enfranchisement, disfranchisement, and reenfranchisement of blacks in Louisiana. The organ of the Court in this case was the organ of the Court in United States v. Louisiana, E.D.La.1963, 225 F.Supp. 353, aff’d 380 U.S. 145, 85 S.Ct. 817, 13 L.Ed.2d 709. In *907that three-judge case, the court, starting with the Code Noir of 1724, examined in exhaustive detail the various methods used in Louisiana to frustrate or to dilute Negro voting, because the court considered it necessary to show that a citizenship test, apparently neutral on its face, was part of a pattern, a member of a long series, designed to prevent or to dilute blacks voting. (See 225 F.Supp. 363-381, 18 pages of a much longer opinion, which on other pages also touched on the same subject.) We summarized the history of Louisiana’s efforts to repress blacks voting as follows:

The Louisiana Codes Noir of Colonial times and the Black Codes of the eighteen sixties; the pre-Civil War denial of the vote to Negroes, even to wealthy and educated free men of col- or; the ebb and flow of Negro rights in the Constitutions of 1864 and 1868; the 1879 transfer of political power from police juries and the legislature to the Governor; the close election of 1892 and the 1896 victory for white supremacy; the grandfather clause and the complicated registration application form in the Constitution of 1898; the invalidity of the grandfather clause and the consequent resort to Mississippi’s understanding and interpretation clause; the effectiveness of the white primary as a means of disfranchising Negroes; the invalidity of the white primary and the consequent need to revive enforcement of the interpretation test; the White League and the Citizens’ Councils; the Black League and the N.A.A.C.P.; the Battle of Liberty Place in 1874 and the Ouachita voting purge of 1956 — these are all related members of a series, all reactions to the same dynamics that produced the interpretation test and speak eloquently of its purpose.

In sum, the interpretation test is another grandfather clause. Its purpose is rooted in the same history. It has the same objective the delegates to the Constitutional Convention of 1898 envisaged for the grandfather clause. It is capable of producing the same effective disfranchisement of Negroes today that the grandfather clause produced sixty-five years ago.

The categorical statement can be made, fairly, that the use of historical boundaries was never resorted to in Louisiana as a means of frustrating or diluting the black vote for members of the legislature. To the shame of Louisiana, among other states, more direct, humiliating, and effective means were used.

On January 1, 1897, thanks to Reconstruction laws, there were in Louisiana 130,344 registered Negro voters and 164,088 white voters.39 There had been many black members of the legislature, a black Lieutenant-Governor, and, for a time, a black Governor.40 Then came the Louisiana Constitution of 1898, a product of the 1896 election victory for white supremacy. The Constitutional Convention of 1897, for fear that the “understanding” clause was vulnerable to constitutional attack, adopted the “grandfather” clause, an “educational” qualification, and a “property” qualification for ■ registration. The results were predictable. In 1900 the following registration figures show that a drastic purge of Negro voters occurred:

White Negro

Under the "educational" qualification 86,157 4,327

Under the "property" qualification 10,793 916

Under the "grandfather" clause 29,189 0

The drop in black registration continued. In 1910 only 730 blacks or less *908than 0.5 percent of the adult male blacks in tin- state were registered. As late as in 1942 only 957 blacks were registered to vote in Louisiana and not one was on the registration rolls in 51 of the 64 parishes in Louisiana. Most of the blacks on the rolls were registered as Republicans. Even in the late fifties and sixties the weapons used to prevent blacks from voting were complicated registration forms, literacy tests (Louisiana has the lowest literacy rate of any state), interpretation tests, and even facially neutral citizenship tests with discretionary power vested in the registrar of voters to evaluate the tests. This is not to minimize the effect of intimidation and economic reprisals as weapons Louisiana whites used against potential black voters. Absent this arsenal of weapons available until only a few years ago, reliance on historical boundaries of legislative seats would have elected many black lawmakers and legislatures would have had the racial composition of the Louisiana legislatures between 1868 and 1896.

In recent years, after blacks began to register freely and in large numbers, city councils, police juries (county commissions), school boards, and other political subdivisions or agencies turned to voting at-large, multi-member districts, anti-single-shot laws difficult for some voters to understand, and perhaps some racial gerrymandering to dilute the black vote. We have none of these devices here.

The Supreme Court’s trust in the district court’s finding of “a long history of bias and franchise dilution in the State’s traditional drawing of district lines” was misplaced.

The trial judge’s leap to his conclusion apparently rests on a single short colloquy he had with Senator Theodore H. Hickey, who shared his district with Senator O’Keefe. It is as follows:

By The Court:

Q. Senator, let me ask you one thing, just so the record will be clear. You refer to the so-called historical boundaries in this area. It is my understanding, and correct me if I am wrong, — but it is my understanding that adherence to these so-called historical boundaries have produced, I believe, since Reconstruction days, two Negro legislators. If that-*-am I correct ?

A. To my knowledge and the terms that I have served, I only remember, Judge, to the best of my memory, two Negro representatives. That is exactly correct.

It seems clear to us that the witness intended no revision of history, leading to any such extraordinarily incorrect historical conclusion as the trial judge drew. Senator Hickey is a veteran of many Louisiana political wars. He simply stated that he could remember only two Negro legislators having been elected, and nothing more. All the rest was in the mind of the questioner, not the answerer. It is a non sequitur to assume that because only two Negroes had been elected to the legislature the producing cause was the state’s policy of adhering to historical political boundaries. Senator Hickey’s answer does not connect the lack of black legislators with legislative districting. He knew better.

VI.

We deferred writing this opinion in part because the Louisiana Legislature had passed Act 457 of 1972, which adopted the Steimel plan as modified by this Court and in part because we were waiting, first, for the decision of the Supreme Court in White v. Regester, 1973, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 and then for two decisions of this Court, Zimmer v. McKeithen, 5 Cir. 1973, 485 F.2d 1297, en banc, reversing 467 F.2d 1381, and Turner v. McKeithen, 5 Cir. 1973, 490 F.2d 191. Unfortunately, the facts and issues in these eases, as developed in the opinions, have not proved especially helpful.

In White v. Regester, as in Whitcomb v. Chavis, the Court held that the plaintiffs must prove some denial of fair and *909effective representation; that they may show, and did show in that case, that they “had less opportunity than did the other residents in the district to participate in the political processes and to elect' legislators of their choice”. 412 U.S. at 766, 93 S.Ct. at 2339. In that case multi-member districts, in conjunction with a variety of state rules and procedures, were employed to exclude blacks and Mexiean-Americans from the electoral process.

Zimmer v. McKeithen involved an at-large voting scheme for electing Police Jurors and School Board members in East Carroll Parish, Louisiana. The district court applied a per se rule that when blacks constitute a majority in a parish, an at-large plan could not dilute their vote. This Court repudiated that rule and observed that “access to the political process and not population was the barometer of dilution of minority voting strength”. 485 F.2d at 1303. Turner v. McKeithen involved a multimember election plan for a police jury, which “dilutes the potential for political participation by the black community of Ouachita Parish”. Chief Judge Brown, writing for the Court, made a comment, pertinent to the instant case:

While a minority group is not constitutionally entitled to an apportionment structure designed to maximize its political advantage, neither may it be enveloped in a structure which will necessarily minimize its potential for meaningful access to the political process.

490 F.2d at 197.

In Wright v. Rockefeller, 1964, 376 U.S. 52, 84 S.Ct. 603, 11 L.Ed.2d 512, the litigation was cast in terms of racial gerrymander. Four districts in Manhattan showed these statistics:

District Percent of Whites Percent of Blacks and Puerto Ricans

17 94.9 5.1

18 13.7 86.3

19 71.5 28.5

20 72.5 27.5

The minority group intervenors defended the concentration of minority voters in the 18th District. The Supreme Court (7 to 2) upheld the districting on the ground that there was no proof that it was racially motivated. Justice Douglas, dissenting, found that the effect of drawing the boundary between two particular districts was “to bring into the Eighteenth District and keep out of the Seventeenth as many Negroes and Puerto Ricans as possible”. (In the instant case the boundary between Steimel’s districts 2 and 3 and that between districts 4 and 5 are drawn so as to bring as many' blacks into districts 2 and 4 as possible and to keep blacks out of districts 3 and 5.) Justice Douglas concluded that “[r] acial segregation that is state-sponsored should be nullified whatever may have been intended”. In the instant case there is certainly no proof that the traditional ward boundaries on which the Senators based their plan were drawn with any intention to discriminate. On the other hand, both the special master and the trial judge candidly conceded that the Steimel districts were drawn in racial terms. We agree with Justice Douglas: “Rotten boroughs were long a curse of democratic processes. Racial boroughs are also at war with democratic standards.”

What Justice Douglas stated so forcefully in his dissent might have been stated as forcefully by Justice Black, for the court, if there had been proof that the districting was racially motivated. Here, the district judge clearly spoke in terms of racial districting. He noted that in his plan two of the four districts would have a majority of black registrants (58 percent and 51 percent), “with the attendant possibility of electing negroes to the Legislature”.

This Court has never been timid in exercising its equitable powers to fashion appropriate color-conscious remedies. “We are not Johnny-come-latelys to the eradication of racial discrimination through race conscious means”, as Chief Judge Brown wrote, concurring, in Morrow v. Crisler, 5 Cir. 1974, 491 F.2d 1053. “The Constitution is both colorblind and color conscious. To avoid conflict with' the equal protection clause, a *910classification that denies a benefit, causes harm, or imposes a burden must not be based on race. In that sense, the Constitution is color blind. But the Constitution is color conscious to prevent discrimination being perpetuated and to undo the effects of past discrimination.” United States v. Jefferson County Board of Education, 5 Cir. 1966, 372 F.2d 836, 876, cert. denied, 389 U.S. 840, 88 S.Ct. 67, 19 L.Ed.2d 103 (1967). In United States v. Louisiana we adopted the freezing principle to enjoin the use of a citizenship test, neutral on its face and as applied, as a voters registration requirement. The object was to give blacks who had been discriminated against in registering a chance to catch up. In school cases, we have imposed ratios as a goal, as to pupils, and as a command, as to teachers.41 In employment cases we have used the race conscious “rightful place theory” 42 and racial hiring quotas.43 In jury selection cases we have approved remedial racial classifications.44

In all these and similar cases the remedy favored integration; had race been used to increase segregation we would never have fashioned the remedies we adopted. Here the Steimel districting tends to segregate racial populations. It may guarantee the election of two black senators but it also guarantees the election of two white senators from districts where there will be few blacks, so few that they may rightfully consider that they have been discriminated against.

VII.

With different pleadings, a different record, and different legal theories, the Taylor plaintiffs might have raised an important constitutional issue that would be closely related to the issue that concerned the Supreme Court. It has been suggested that “we are leaving the era of reapportionment and beginning the quest for representative apportionment”. 45 Foreseeing that day, one commentator has stated the issue in these terms: “The issue is whether it is constitutionally permissible to use race explicitly as a criterion in reapportionment, when the goal of racial districting is purportedly benign”.46 Mr. Stanley Halpin, counsel for Mrs. Taylor, in a recent article, concluded, however, that thus far the Supreme Court “has not indicated any inclination to sustain a racial gerrymandering claim”. Halpin and Engstrom, “Racial Gerrymandering and Southern State Legislative Redistricting”, 22 Jour.Pub.Law 37, 41 (1973).

The question raises so many complex political and social problems as well as constitutional difficulties, it is no wonder that neither counsel for Mrs. Taylor nor Mr. Steimel nor the district court was willing to accept that question as the issue in this case, with the Steimel plan as the answer to the question. Nor can we say that on the record before us, that was the issue. Not the least of the problems that would be created by representative apportionment is that the *911cartographer, whether he be a legislative commissioner or a court-appointed master, cannot draw his meandering lines without increasing segregation and aggravating the polarization of the races in the community. Here, for example, Steimel could not pack a black population of 70.2 percent into his district 4, except by cutting down the number of blacks in his district 3 to 16 percent. Mrs. Taylor might have argued that representative apportionment legitimated the gerrymander. She did not resort to this argument.

No doubt a strong case can be made for the use of purposeful judicial racial gerrymandering to afford blacks fair representation in the legislature, substantially proportionate to their population or at least to their registration.47 This argument could have been made here even though the Senators’ plan and, for that matter, the traditional boundaries of voting districts were not “rooted in racial discrimination”. The district court did not see the case that way. Instead, the district court based its decision on the clearly erroneous notion that Louisiana history demonstrated that blacks were deprived of access to the legislature by discriminatory districting. On the merits of the two plans, the district court’s preference for the Steimel gerrymander of districts 2, 3, 4, and 5 exceeded the bounds of judicial discretion. As we see it, the alternative non-gerrymandered plan of the Senators was not one composed of traditional “safe” white districts. Although based on traditional ward lines, it was radically different from all of the State’s previous statutory apportionments of the area in question, geographically and in having all single-member districts. Considering the totality of circumstances relating to the four districts as a whole, the Senators’ plan accords blacks more access and heavier weight in the political processes than the plan approved by the district court. This is especially evident if we look to population trends. In short, sound judicial discretion commands approval of the Senators’ plan.

The foregoing opinion explains the reasons for our original decision, and our belief that this judgment was correct. Decisions since the date of our judgment either support that judgment or are not inconsistent with it. Accordingly, with all due respect and deference to the Supreme Court, we reinstate our original judgment.48

Appendix to follow.

*912

*913SEIMEL PLAN