Andrew J. McGHEE; Lacy Joyner; Darryl Moss; Eugene Fields; Aurelia B. Burton, on behalf of themselves and all others similarly situated, Plaintiffs-Appellees, v. GRANVILLE COUNTY, NORTH CAROLINA; Granville County Board of Commissioners; Samuel W. Daniel, Chairman of the Granville County Board of County Commissioners; W.E. Averette; J. David Brooks; L. Sam Daniel; R. Wayne Newton, Commissioners of Granville County, North Carolina and their successors and agents, Defendants-Appellants and Granville County Board of Elections; Franklin E. Elliott, Jr., Chairman of the Granville County Board of Elections; Helen Amis; Karen W. Brown, Members of the Granville County Board of Elections, and their successors and agents; Cindy Burnett, Supervisor of the Granville County Board of Elections, Defendants, 860 F.2d 110

JAMES DICKSON PHILLIPS, Circuit Judge: This appeal presents the issue of the extent of a federal court’s remedial power in reviewing a legislative plan designed, in response to the court’s order, to remedy a stipulated violation of

Section 2 of the Voting Rights Act of 1965, as amended. 42 U.S.C. § 1973 (as amended). Based upon a stipulation that the challenged at-large method of electing members of the Gran-ville County, North Carolina, Board of County Commissioners violated § 2, the district court ordered the parties to attempt to agree upon a remedy, failing which the County was ordered to submit a proposed remedial plan.

After the parties failed to agree, the court rejected the County’s proposed single member district plan and instead ordered into effect a modified version of the complaining parties’ responsive proposal for a plan based upon “limited voting” in at-large elections. Because we conclude that the district court erred in rejecting the County’s plan, we reverse and remand for implementation of the County’s proposed remedial plan.

I This action was commenced on January 15, 1987, by five black citizens and registered voters of Granville County, North Carolina, on behalf of themselves and all other black voters of the county against the County, the County Board of Commissioners, its members, the County Board of Elections, its members, and the County Supervisor of Elections.

The plaintiffs alleged that the then existing at-large method of electing the Granville County Board of County Commissioners (the Board) had the result of “diluting minority voting strength and denying members of the black community the opportunity to elect representatives of their choice” to the Board, in violation of § 2 of the Voting Rights Act of 1965, as amended. 42 U.S.C. § 1973 (§ 2, or the Act). The Board is the governing body of Granville County.

At the time the plaintiffs brought this action, the Board consisted of five members, on a county-wide at-large basis, but required to reside in particular residence districts. Each member was elected for a four-year term. The terms of the various Board members were staggered, with elections being held in even numbered years. Three of the five incumbent members were serving terms which expired in 1988. The remaining two incumbent members were serving terms which expire in 1990.

Black citizens make up 43.9% of the county’s total population (1980 data), 40.8% of its voting age population (1980 data), and 39.5% of its registered voters (1987 data). Despite these population numbers, and despite the fact that a number of black residents have run for election to the Board, no black has ever been elected to the Board.

On the parties’ joint pre-trial motion, in which the County stipulated that the challenged electoral scheme “does not comply with the requirements of § 2 of the Voting Rights Act,” the district court entered a consent order which required the parties to attempt to agree upon a remedial plan, failing which the county would submit a proposed remedial plan, to which plaintiffs might submit a response for consideration by the court.

In accordance with the order, after the parties failed to agree upon a remedy, the County submitted its proposal, which had earlier been given § 5 preclearance by the Attorney General of the United States. See 42 U.S.C. § 1973c. The County’s proposal was for a single member district electoral plan containing seven districts, with members serving staggered terms, thereby both abandoning the at-large election method and expanding Board membership from five to seven.

The districts proposed contained the following black population percentages (voting age figures as estimated under district court formula): DISTRICT TOTAL BLACK POPULATION (%) BLACK VOTING AGE POPULATION (%) 1. South Oxford/Fishing Creek 70.3 67.5 2. Oxford/North Oxford 42.6 39.5 3. Sassafras Fork/Salem/East Oxford 36.4 33.5 4. Dutchville, West of 1-85 31.3 28.6 5. Brassfield/Dutchville, East of 1-85 33.5 30.7 6. Oak Hill/Walnut Grove/Tally Ho/South Fork 55.0 51.8 7.

Tally Ho/Brassfield/Fishing Creek 39.8 36.8 In their response to the County’s proposed remedial plan, the plaintiffs did not contend that its districting feature failed to provide the maximum remedial relief possible by that means. Indeed, they conceded then and continue on this appeal to concede that “due to the demographics of Granville County, it is not possible to draw a five or seven single member district plan that gives black voters any better opportunity to elect representatives of their choice.” Appellees’ Brief at 4 n. 2.

Rather, plaintiffs contended that single member district-ing was shown by the plan’s demographics to be inadequate as a remedial device in Granville County. The very best plan possible could do no more than provide one “safe” district (District # 1) and one in which there was no better than a fighting chance (District # 6). Whereas overall black voting age population in the County was 40.8%, single member districting could give blacks no more than 14-28% representation on a seven member Board.

The plaintiffs’ objection to the County’s proposed remedial plan was explicit: it would not provide black citizens “a chance to elect a number of commissioners that is commensurate with their portion of the population and with their voting strength.” Plaintiffs’ Memorandum in Support of Alternative Remedies at 4. As an alternative, plaintiffs proposed a “limited voting plan.” Under plaintiffs’ proposal, the Board would be composed of seven members elected concurrently on a county wide at-large basis, with voters allowed to vote for any three or fewer candidates as they chose.

Plaintiffs asserted to the court that this plan would give black voters a fair chance of electing three commissioners, 42% of the Board. After considering both plans, the district court rejected the County’s plan.

The court was explicit: The County’s plan “does not 'completely’ remedy the existing dilution of black voting strength in Gran-ville County or provide plaintiffs with an equal opportunity to participate in the political process.” The plan was inadequate as a remedy because it gave black voters, some 41% of the voting population, little likelihood of electing more than two commissioners (28% of the board), and more likely gave them a chance to elect only one commissioner (14%). Having rejected the County’s plan on this reasoning, the court adopted a modified version of the plaintiffs’ limited voting plan.

The court’s plan was specific and detailed: 1. The Board shall be expanded to seven members at the 1988 election, and shall remain at that number indefinitely. All terms shall be for four years except as otherwise stated in this order. In all elections, the candidates who receive the highest number of votes shall be deemed elected. 2. The two Commissioners not up for re-election in 1988 shall retain their seats until their terms expire in 1990. 8. In all elections, except as specifically noted herein, voters shall be limited to voting for a maximum of two candidates to fill the available seats. 4.

Five Commissioners shall be elected in 1988, with the Commissioner receiving the fewest votes to serve a two-year term to expire in 1990. 5. Starting in 1990, an election shall be had every four years, in even years not divisible by four, to fill the three expiring terms; starting in 1992, an election shall be had every four years in even years divisible by four, to fill the other four expiring terms. 6. Beginning with the election in the year 2000, and in all elections and subsequently in even years divisible by four, voting shall be at large.

In subsequent elections, in years not divisible by four, voters shall continue to be limited to voting for a maximum of two candidates to fill the available seats. This appeal by the County followed. II The dispositive issue, put in broadest terms, is whether the district court properly could reject the County’s remedial single-member district plan and impose instead its own modified version of the plaintiffs’ limited voting plan.

To address that issue we first summarize the controlling principles respecting judicial review of legislative plans submitted, in obedience to court decrees, to remedy judicially established violations of § 2 of the Voting Rights Act. Where, as here, a court has properly given the appropriate legislative body the first opportunity to devise an acceptable remedial plan, see White v. Weiser, 412 U.S. 783, 794-95, 93 S.Ct. 2348, 2354-55, 37 L.Ed.2d 335 (1973), the court’s ensuing review and remedial powers are largely dictated by the legislative body’s response.

If the legislative body fails to respond or responds with a legally unacceptable remedy, “the responsibility falls on the District Court,” Chapman v. Meier, 420 U.S. 1, 27, 95 S.Ct. 751, 766, 42 L.Ed.2d 766 (1975) (reapportionment case) to exercise its discretion in fashioning a “near optimal” plan. Seastrunk v. Burns, 772 F.2d 143, 151 (5th Cir.1985) (same).

Where, however, the legislative body does respond with a proposed remedy, a court may not thereupon simply substitute its judgment of a more equitable remedy for that of the legislative body; it may only consider whether the proffered remedial plan is legally unacceptable because it violates anew constitutional or statutory voting rights — that is, whether it fails to meet the same standards applicable to an original challenge of a legislative plan in place. Upham v. Seamon, 456 U.S. 37, 42, 102 S.Ct. 1518, 1521, 71 L.Ed.2d 725 (1982).

If the remedial plan meets those standards, a reviewing court must then accord great deference to legislative judgments about the exact nature and scope of the proposed remedy, reflecting as it will a variety of political judgments about the dynamics of an overall electoral process that rightly pertain to the legislative prerogative of the state and its subdivisions. See Weiser, 412 U.S. at 795, 93 S.Ct. at 2354; Cook v. Luckett, 735 F.2d 912, 920-21 (5th Cir.1984) (district court’s rejection of legislative proposal in § 2 case reversed).

Applying those principles, we are satisfied that the County’s remedial plan here met the relevant standards and that the district court therefore erred in declining to accept it as a “complete” remedy for the specific violation of § 2 voting rights that had been alleged and established. To show why requires analysis of the specific violation claimed and found and of the remedy for it that was proposed by the County. Ill We start with the slight awkwardness that the violation here was stipulated pretrial, not adjudicated on a full evidentiary record, and that it was framed in general rather than specific terms.

It was simply, as indicated, that the challenged electoral process “does not comply with § 2 of the Voting Rights Act.” Despite the generality of the stipulation, resort to the record, particularly the pleadings, makes clear the specific nature of the violation first claimed by plaintiffs, then stipulated by the County, and finally considered by the district court to have been judicially established.

As there revealed, the specific violation alleged and established by stipulation and consent decree was classically one of “vote dilution” by the “submergence” of minority voters’ potential voting power through the use of an at-large electoral process.

Because the nature of such a voting rights violation bears critically upon our decision, we first consider the basic “vote dilution” concept as it has now evolved in statutory and decisional law. “Vote dilution” resulting, in varying forms, from geographical districting practices is but one of various ways in which the voting rights of racial minorities may be “denied or abridged” in violation of § 2.

But it is a distinctive way which in recent times, with the gradual eradication of most of the more direct historical forms of outright denials or subversions of voting power and access itself — literacy tests, poll taxes, anti-single shot laws and the like — has become the dominant remaining means of voting rights violations, and hence the most frequent focus of contemporary legal challenges.

First originating as a judicially recognized special form of voting rights violation, the “vote dilution” concept, stripped of any requirement of discriminatory intent, has now of course been effectively codified as a special form of violation, in amended § 2. See Thornburg v. Gingles, 478 U.S. 30, 42-51, 106 S.Ct. 2752, 2762-67, 92 L.Ed.2d 25 (1986).

Being less obvious and direct a means by which § 2 voting rights may be violated than the more brutally direct formal devices now largely of the past, the concept underlying vote dilution is correspondingly more subtle and difficult to keep within principled legal bounds — both in the violation and remedial stages of applying § 2. A moment’s reflection — confined to the dilution-by-submergence concept here in specific issue — shows why.

The basic concept, broadly stated, is that racial minorities may not have their group voting power impermissibly “diluted” by multimember districting or at-large electoral processes which “submerge” the minority voting group in a voting constituency in which the voting power of a racially “bloc-voting” white majority always insures defeat for the candidates of the minority group’s choice. Gingles, 478 U.S. at 46, 106 S.Ct. at 2764. As so stated, the concept is logically unbounded.

It has no implicit limits related to any of its principal components: how racial majority and minority voters are geographically dispersed in the overall voting constituency; their respective population percentages; whether and to what extent the two racial groups vote along racial lines; and the number of representatives involved. All of these are essential functions of the ultimate phenomenon of voting power “dilution” as that term has acquired the legal meaning above summarized.

Except as these elements are given bounds, therefore, the concept is simply an open-ended one subject to no principled means of application. Ultimately, unbounded, it could be applied to find “dilution” of a minority group’s voting power in any situation where the group had been unable, despite effort, to achieve representation by the election of candidates of its choice in proportion to its percentage of the total voting age constituency.

Courts applying the concept have, of course, always realized the problems presented by its theoretically open-ended nature, From the outset they have responded by imposing special requirements upon its constituent elements, such as, for example, upon the minimum size and characteristics of a racial minority group that could) be considered an “effective voting majority,” both for purposes of finding that “dilution” of such a group’s potential voting power existed and for then devising an appropriate districting remedy. See, e.g., McNeil v.

Springfield Park Dist., 851 F.2d 937, 944-45 (7th Cir.1988) (reviewing courts’ requirements). The same concern to keep the concept within principled bounds was of course dominant in congressional deliberation leading to its codification, now stripped of any intent requirement, in amended § 2. The concern, specifically addressed, was that without adequate legal constraints upon judicial discretion to find and remedy mere vote dilution “results,” the inevitable consequence would be a proportional representation approach.

The response, upon which it is generally considered passage finally turned, was the addition to § 2(

b) of the “Dole Compromise” proviso, which specifically disclaimed any legislative intent to establish any “right” of proportional representation. See note 1, supra; see generally Gingles, 478 U.S. at 84, 106 S.Ct. at 2784 (O’Connor, J., concurring). The judicial and legislative process of putting principled bounds upon the vote dilution concept has now culminated in the Supreme Court’s exhaustive analysis of the concept as codified in amended § 2, in Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986).

After specifically recognizing that the dilution concept, stripped of any discriminatory intent requirement, had been deliberately codified by Congress in the amended Act, the Gingles Court proceeded to find in it three threshold proof requirements.

Specifically, the minority group must prove, as “preconditions,” that (1) “it is sufficiently large and geographically compact to constitute a majority in a single-member district”; (2) that “it is politically cohesive”; and (3) that “the white majority votes sufficiently as a bloc to enable it — in the absence of special circumstances, such as the minority candidate running unopposed ..., usually to defeat the minority’s preferred candidate.” Gingles, 478 U.S. at 50-51 & nn. 16-17, 106 S.Ct. at 2766-67 & nn. 16-17; see also Collins v.

City of Norfolk, Virginia, 816 F.2d 932, 935 (4th Cir.1987) (recognizing Gingles preconditions as “essential” to proof of vote dilution); Campos v. City of Baytown, 840 F.2d 1240, 1244 (5th Cir.1988) (“Gingles threshold” factors recognized as prerequisites to proof of dilution claim); McNeil, 851 F.2d at 942 (same; establishment of preconditions essential to “pass the

summary judgment threshold”); Cruz Gomez v. City of Watsonville, 852 F.2d 1186, 1191-92 (9th Cir.1988) (Gingles factors “prerequisites” to proof of causal connection between districting system and dilution of voting power). In addition to identifying these essential elements of a vote dilution claim, the Gin-gles Court, elaborating its analysis, made several other observations about the nature of this claim that are critical to our decision.

First, emphasizing the centrality of the “size and compactness” element, the Court pointed out that where compactness cannot be shown, and where, instead, “minority voters’ residences are substantially integrated throughout the jurisdiction, the at-large district cannot be blamed for the defeat of minority-supported candidates.” Gingles, 478 U.S. at 51 n. 17, 106 S.Ct. at 2766 n. 17 (quoting Blacksher & Menafee, From Reynolds v. Sims to City of Mobile v. Bolden: Have the White Suburbs Commandeered the Fifteenth Amendment?, 34 Hastings L.J. 1, 56 (1982)).

Further, and of particular importance to our analysis, the Court noted that the “size and compactness” requirement confines dilution claims to situations where diminution of voting power is “proximately caused by the districting plan,” id., and thus “would not assure racial minorities proportional representation.” Id. (emphasis the Court’s).

Finally, in defining the “essence of a § 2 claim,” the Court made plain that the adverse “result” for which such a claim seeks a remedy must be traceable ultimately to the impact of “a certain electoral law, practice or structure interacting] with social and historical conditions.” Id. at 47. IV From these developments respecting the vote dilution concept and § 2 claims based upon it, particularly the Supreme Court’s Gingles analysis, we deduce the following interrelated principles critical to decision here. 1.

The “certain electoral law, practice, or structure” necessarily challenged as an essential element of a § 2 vote dilution claim is the districting system which allegedly “submerges” (or “fractures” or “packs”) the minority group’s putative voting power. 2.

If a vote dilution violation is established, the appropriate remedy is to restructure the districting system to eradicate, to the maximum extent possible by that means, the dilution proximately caused by that system; it is not to eradicate the dilution by altering other “electoral laws, practices, and structures” that were not actually challenged by the claim as made. 3. The maximum extent to which a particular dilution violation may be remedied by restructuring the districting system is constrained by the size, compactness, and cohesion elements of the dilution concept.

A restructuring to the maximum extent permitted by these constraints is a “complete” and legally adequate remedy for such a dilution violation. 4. The disclaimer in amended § 2 of any “right” of racial minorities to proportional representation prevents a court from using proportional representation as the ultimate standard for assessing the legal adequacy of a remedial legislative redistricting plan. If these principles be accepted — as we do — they of course reveal the fallacy of the plaintiffs’ position and the error of the district court’s rejection of the County’s proposed remedial plan.

Within those principles, the plaintiffs’ concession that the County plan provided the maximum remedy possible by redistricting establishes the plan as a legally adequate one that should have been accepted in deference to the affected local government’s primary jurisdiction to ordain its electoral processes. The plaintiffs’ principal contentions to the contrary deserve some discussion. A Their primary contention, expressly accepted by the district court, is that the districting remedy, though concededly maximum by that means, was not the “complete” one legally required.

Though not made explicit, it is obvious that this view of “completeness” of remedy has to assume one of two legally erroneous standards against which to measure “completeness.” The first standard necessarily looks to the plight of those minority voters not included within one of the remedial plan’s “safe” districts. As to those included within those districts, the remedy is manifestly “complete,” both legally and practically. It is only as to those voters not included that the plan might be considered in any sense an incomplete eradication of the “submergence” caused by the at-large voting system.

For some, possibly all of these voters, the remedial plan “submerges” their voting power — at least in direct terms — to a greater degree in their new constituencies than did the challenged system in the old. This, of course, is simply an unavoidable mathematical consequence of the demographics that will constrain practically all single-member district remedial plans. It is undoubtedly to some extent an ironic consequence. But it is a consequence that has not been thought to invalidate remedial plans by those courts that have considered the matter directly.

Those courts have instead, apparently without exception, simply accepted it as a necessary concomitant of the inevitably rough-hewn, approximate redistricting remedy. See Cruz Gomez, 852 F.2d at 1193 (“[districting plans with some members of the minority group outside the minority-controlled districts are valid”); Campos, 840 F.2d at 1244 (“The fact that there are members of the minority group outside the minority district is immaterial.”); see also Gingles v. Edmisten, 590 F.Supp. 345, 380-84 (E.D.N.C.1984) (three-judge court), aff'd in part, rev’d in part on other grounds, sub nom., Thornburg v.

Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). As our analysis of Gingles indicates, we believe that the Supreme Court was fully aware of this particular aspect of the re-districting remedy even as it necessarily implied that it was the appropriate and adequate remedy despite this possible short-fall effect.

We agree with Judge Cu-dahy’s recent observation that the Gingles Court’s careful effort to contain the vote dilution concept and claims based upon it within principled bounds necessarily involved a deliberate “trade-off” which “precludes some small and unconcentrated minority groups from attempting to rectify vote dilution.” McNeil, 851 F.2d at 942 (Gingles approach “reins in the almost unbridled discretion that

Section 2 gives the courts”); see also Gingles, 478 U.S. at 84, 106 S.Ct. at 2784 (O’Connor, J., concurring in judgment) (emphasizing “compromise” involved in § 2’s enactment). Consequently, we agree with the Fifth and Ninth Circuits that a remedial districting plan is not invalidated — made “incomplete” — solely by the consequence that some of the racial minority group remains “submerged” in non-safe districts. This is not an appropriate standard for assessing the adequacy of a remedial districting plan.

The other standard implicit in plaintiffs’ contention, and in the district court’s reasoning, is even more plainly inappropriate. Despite the disavowals of both, it is obvious that ultimately the plaintiffs urged and the district court accepted a proportional representation standard. This directly violates the § 2(

b) proviso expressly disclaiming any such “right.” Plaintiffs apparently seek to avoid the proviso’s effect, as we understand their position, by insisting that vote dilution violation and vote dilution remedy are to be separately considered in assessing the proviso’s intended effect.

On this view, the proviso only prohibits finding a violation based solely upon a lack of proportional representation; where a violation properly traceable to a specific voting mechanism is found, the proviso does not then prohibit a court from rejecting a proposed remedy that does not assure approximate proportional representation and imposing one that does. Here the violation established was vote dilution specifically caused by an at-large electoral system, not simply the lack of proportional representation.

That exhausts the force of the proviso, and the court could then properly find invalid a proposed legislative remedy which fell short of assuring approximate proportional representation and substitute one of its own which did. This contention seizes upon and seeks to exploit the undoubted logical problem created by Congress’ simultaneous allowance of vote dilution claims, which necessarily require some consideration of population proportions, and the disclaimer of any “right” to proportional representation.

See Gingles, 478 U.S. at 84, 106 S.Ct. at 2784 (O’Connor, J., concurring in judgment) (pointing to “inherent tension between what Congress wished to do and what it wished to avoid”). The problem of statutory

interpretation is certainly there, but we are satisfied that to adopt the plaintiffs’ position and affirm the district court’s reasoning and judgment would simply negate the proviso, and defeat Congress’ intention in adopting this disclaimer as the ultimate back-stopping principle of this “compromise” legislation. See id. The practical consequence of uncoupling violation from remedy in this way would necessarily be to allow proportional representation to become in practical effect the “right” protected by § 2. Ubi jus, ibi remedium, and vice versa.

Certainly implicit in the Gingles Court’s analysis of the nature of § 2 vote dilution claims is the notion that, so far as those claims are concerned, right and remedy are inextricably bound together, for to prove vote dilution by districting one must prove the specific way in which dilution may be remedied by redistricting. See McNeil, 851 F.2d at 942 (“Court’s approach ... focusing up front on whether there is an effective remedy for the claimed injury ... reins in the almost unbridled discretion that

section 2 gives courts”). Whatever its other effects, we therefore believe that the § 2 proviso prevents a court from rejecting a remedial legislative districting plan which provides the maximum opportunity for representation possible by that means for the sole reason that the representation possible does not sufficiently approximate proportionality. B By way of justifying the district court’s rejection of the County’s plan in favor of its demonstrably more effective “limited voting” plan, the plaintiffs seek to establish the validity and the virtue of limited voting as a viable electoral process.

They point out that it has been used in some American jurisdictions by free legislative choice; that it has withstood equal protection and other constitutional attacks; and that in fact it has been adopted by consent decrees as a remedial system in at least three recent § 2 vote dilution cases. The short answer to all these examples of the device’s use and acceptance is that they are irrelevant to the issue before us. In the first place, the specific issue here is not the validity vel non of the district court’s substituted plan, but the prior adequacy of the County’s plan.

Hence the fact that freely chosen legislative plans incorporating this feature might pass constitutional muster under equal protection or other attack is irrelevant. So too is the fact that some legislative bodies may have entered into consent decrees adopting the system as part of a remedial plan for a § 2 violation.

Section 2 only disclaims any judicially enforceable federal “right” to proportional representation; it nowhere prevents judicial approval of electoral plans specifically designed by responsible legislative bodies to assure proportional representation as a specific remedy for a § 2 violation. See United Jewish Orgs. v. Carey, 430 U.S. 144, 168, 97 S.Ct. 996, 1011, 51 L.Ed.2d 229 (1977). V We therefore must reverse the district court’s judgment and remand for entry of an appropriate order approving and implementing the County’s proposed remedial plan.

In view of the fact that the district court’s remedial plan has already been put into effect and implemented through the primary election stage, we realize that major problems are presented for undoing what has been done and putting the County’s plan belatedly into effect. Because the next critical stage is the impending general election scheduled for November 8, 1988, we realize that the first steps must be taken expeditiously to avoid further disruption.

We will therefore direct that the mandate be issued forthwith so that the district court may address the immediate problem of the upcoming general election as a prelude to direct implementation of the County’s plan. Without intending to limit the district court’s discretion in deciding how best at this juncture to implement the County’s plan, we observe that there are two basic alternatives, either of which would find support in precedent.

The first would cancel the 1988 primary results and enjoin the pending general election, provide that the current members of the Board remain in office until successors were elected under the County plan, and

schedule a special primary and general election under that plan. See City of Richmond v. United States, 422 U.S. 358, 365, 95 S.Ct. 2296, 2301, 45 L.Ed.2d 245 (1975). The second would permit the pending general election to proceed, provide that members then elected might serve until successors were elected under the County plan, and

schedule a special primary and general election under that plan. See Cosner v. Dalton, 522 F.Supp. 350, 364 (E.D.Va.1981) (three-judge court). SO ORDERED. .

Section 2, as amended in 1982, reads as follows: (

a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in

section 1973b(f)(2) of this title, as provided in subsection (

b) of this section. (

b) A violation of subsection (

a) of this

section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (

a) of this

section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this

section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population. 42 U.S.C. § 1973. Plaintiffs also alleged a constitutional claim against the at-large election system, but did not pursue it after the § 2 violation was stipulated. . After being nominated in primaries subject to a majority-vote rule. N.C.Gen.Stat. § 163-111. .Limited voting is a "semi-proportional" representational system. Under a limited voting plan, elections are held on an at-large basis.

Voters are only allowed to cast a number of votes that is less than the number of seats up for election and, if allowed more than one vote, can choose as well not to exercise all of them. The candidates with the most votes, up to the number of open seats, win. The unassailable theory behind limited voting is that so long as a disproportionate number of minority candidates does not run, even an insubstantial minority group may elect at least one candidate.

The actual number of course will depend on the number of majority and minority candidates and the relative proportions of the two "bloc voting” groups in the voting electorate. For simple example, if in an election for four seats, voters are limited to one vote, a disciplined minority of slightly more than 20% can elect one representative "of its choice," thereby gaining 25% "representation.” See generally Lijphart, Introduction to

Part II: Election Mechanisms Other Than Single Member Plurality Districts, in REPRESENTATION AND REDISTRICTING ISSUES 103, 103-04 (B. Grof-man, A. Lijphart, R. McKay, & H. Scarrow eds. 1982). . Within its premises, the district court's plan is an admirably thoughtful and creative one. By rejecting the plaintiffs’ concurrent-election proposal and continuing the extant staggered-term feature, the court deliberately sought to guard against massive board turnovers during what inevitably was to be a sensitive transition period.

By scaling down the limited voting plan's impact over time, the court sought to take into account an anticipated change in racial polarization voting patterns which the court thought would occur as black representatives allayed traditional white voter concerns about fitness and voting propensities. . After a stay pending appeal was denied both by the district court and then by a panel of this court, voting pursuant to the district court's remedial plan occurred in the Democratic primary election on May 3, 1988. Of the five seats open, black candidates won nomination to three.

We are advised that the five Democratic candidates are unopposed in the pending general election on November 8, 1988. . The critical § 2 allegation of violation in the complaint was that "Defendants' method of electing the Board of Commissioners by an at large election method with residence districts and for staggered terms has the result of abridging plaintiffs’ right to vote ... etc.” The corresponding prayer for relief was for a declaration that "the use of at large, staggered term elections from residence districts ... violates Sec. 2.” Defendants stipulated that "the current method of electing the ...

Board of County Commissioners, as [pleaded in] the complaint ... does not comply with the requirement of § 2____” The district court accepted the stipulation as defining the violation. . Districting may dilute racial minority voting power by "submerging,” "fracturing," or "packing” putative voting age majorities in the various ways suggested by these terms. See generally Derfner, Racial Discrimination and the Right to Vote, 26 Vand.L.Rev. 523, 553 (1973). The claim here is of dilution by "submergence.” .

The district court found the “complete” remedy requirement — which it obviously thought dis-positive — in a passage of amended § 2’s legislative history. The passage, with emphasis added by the court, was quoted in the court’s order rejecting the County’s plan: The court should exercise its traditional equitable powers to fashion the relief so that it completely remedies the prior dilution of minority voting strength and fully provides equal opportunity for minority citizens to participate and to elect candidates of their choice. 1982 U.S.Code Cong, at 208 (emphasis added). J.A. at 43.

This passage of legislative commentary of course does not purport to define the standard against which “completeness” of the remedy is to be measured. . It is to be kept in mind that here — as in dilution cases generally — the claim of dilution by submergence is made by a class consisting of all the black voters of the jurisdiction whose at-large electoral system is challenged. Inherent in any redistricting remedy available to such a class is the possibility — which must be considered one accepted from the outset by the class — that not all can be placed in safe districts. See Gingles v.

Thornburg, 590 F.Supp. at 380-84. That not all can be given the remedy of assured direct representation by a candidate of their choice does not, however, mean that the remedy is wholly ineffectual as to those not included in "safe” districts. Once the concept of racial group “interest” representation is accepted — as surely it has been in the codified dilution concept — it must in logic be assumed that the special interests of minority voters not included in safe districts will nevertheless be “represented,” albeit less directly, by those minority candidates elected from “safe" districts. .

Whether other elements of the County’s remedial plan may now or in time cause different forms of cognizable § 2 harm to these persons as a discrete sub-group of the original plaintiff class is not before us. We only determine here that as to the class of which they are members in this litigation the County plan adequately remedies the dilution-by-submergence violation specifically alleged and established in respect of that entire class.

The impact upon this discrete group of minority-race voters of other elements of the single-member districting plan could only be the subject of dubious judicial prediction and intuition. See Dillard v.

Crenshaw County, 831 F.2d 246, 250 (11th Cir.1987) ("evidence showing a violation in an existing election scheme may not be completely coextensive with a proposed alternative”) (emphasis the court’s); Gingles v, Edmisten, 590 F.Supp. at 380-84 (rejecting claim that remedial plan insufficiently remedied "dilution” for some minority voters not included in "safe” districts; effect of remedial plan upon them too speculative to assess on extant record). .

This might of course be accomplished by other, less exotic, means than "limited voting,” such as increasing the size of the representative body, hence the number of single-member districts, a process that obviously can be used to move toward greater proportionality. See, e.g., Carrollton Branch of NAACP v. Stallings, 829 F.2d 1547 (11th Cir.1987) (direct attack, claiming "dilution," on single commissioner form of government; increase to multiple-member commission proposed as remedy).

Indeed, the County’s remedial plan here incorporated such a further representational enhancing device, a matter clearly within its legislative prerogative. We have no occasion here to consider the extent to which this feature of an electoral plan might be the direct object of a § 2 challenge. It was not the object here. The point is only that by one means or another, if dilution remedy is uncoupled from specific dilution violation, proportional representation inevitably will suggest itself as the appropriate remedy.