OPINION
PER CURIAM.This case is currently before the Court on the following motions: (1) Defendants’ Motion for Summary Judgment on all Counts in Civil Action No. S-92-510; (2) Plaintiffs’ Cross-Motion for Summary Judgment on Count I (“one person, one vote” violations) in Civil Action No. S-92-510; and (3) Defendants’ Motion for Summary Judgment on all Counts in Civil Action No. S-92-1409. A hearing on all motions was held on November 19, 1993. The Court determined that a trial was necessary to resolve factual disputes concerning a potential violation of the Voting Rights Act on the State’s Eastern Shore. The trial was held on December 20 and 21, 1993; and the findings of fact and conclusions of law are set forth below, in *1029accordance with Federal Rule of Civil Procedure 52(a). See infra Part V.C.1 As to all other claims in the combined lawsuits, the Court finds that there is no genuine issue as to any material fact and that the defendants are entitled to judgment as a matter of law. See Fed.R.Civ.P. 56(c).
I. COMMON FACTUAL BACKGROUND
Pursuant to Article III, § 5 of the Maryland Constitution, Governor William Donald Schaefer, following the 1990 federal census, undertook to develop a redistricting plan for the Maryland General Assembly that reflected the population shifts throughout the State since 1980. The Governor appointed a five-member advisory committee, the Governor’s Redistricting Advisory Committee (“GRAC”),2 to assist him in that task in May 1991. After acknowledging the legal criteria that would constrain its redistricting decisions 3 and the non-legal factors that it would also take into account in the redistricting process,4 the GRAC held a series of public hearings during the summer of 1991 across the State to receive public comments on the process and proposals for alternative district-ing plans. After an enthusiastic public response,5 the GRAC worked throughout the fall of 1991 to formulate a proposed redistricting plan which was released to the public on December 2, 1991. The GRAC held a final public hearing on the proposed plan on December 106 and, after several changes made in light of testimony at that hearing, submitted the plan to the Governor on December 17. The Governor submitted a slightly modified version of the plan to the General Assembly on January 8, 1992. That plan became law on February 23, upon the General Assembly’s failure to enact its own redistricting plan.7
This consolidated case involves challenges by two groups of plaintiffs to Maryland’s redistricting plan. The defendants are the Governor, the State Administrative Board of Election Laws and its Administrator, and the Secretary of State. In Civil Action No. S-92-510, the plaintiffs, Marylanders for Fair Representation, Inc. (“MFR”) and two Republican registered voters, claim that the plan violates the “one person, one vote” requirement of the Fourteenth Amendment, is an unconstitutional political gerrymander, and violates the Voting Rights Act. In Civil Action No. S-92-1409, the plaintiffs, the National Association for the Advancement of Colored People, Inc. (“NAACP”), seven of its constituent branches, and eight black registered voters, claim that the plan violates the Voting Rights Act and the Fourteenth and Fifteenth Amendments. The Court of Appeals of Maryland has already upheld the plan against attacks by other plaintiffs based on eight separate state and federal grounds. See Legislative Redistricting Cases, 331 Md. 574, 629 A.2d 646 (1993). Both federal suits have been consolidated to be heard by this three-judge Court. See 28 U.S.C. § 2284.
II. SUMMARY JUDGMENT STANDARD
In a motion for summary judgment, the burden is on the moving party to demon*1030strate by a properly supported motion that there is no genuine dispute as to any material fact and that the movant is entitled to a judgment as a matter of law. Fed.R.Civ.P. 56(c). Because under Rule 56(a) and (b), both plaintiffs and defendants may move for summary judgment, courts are often confronted with cross-motions. In such situations, the court must consider each party’s motion individually to determine if that party has satisfied the summary judgment standard. See 10A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 2720 (2d ed. 1983).
The moving party has the initial responsibility of informing the court of the basis for the belief that summary judgment is warranted. See Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 2552-53, 91 L.Ed.2d 265 (1986); Pulliam Inv. Co. v. Cameo Properties, 810 F.2d 1282, 1286 (4th Cir.1987). If the moving party does not bear the ultimate burden of persuasion, it must show that there is an absence of evidence to support the nonmoving party’s claim. See Celotex, 477 U.S. at 325, 106 S.Ct. at 2553-54. Once a motion for summary judgment is made and supported, the nonmoving party “may not rest upon the mere allegations or denials of [that] party’s pleading, but ... must set forth specific facts showing that there is a genuine issue for trial.” Fed.R.Civ.P. 56(e); see also Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 2510, 91 L.Ed.2d 202 (1986). The nonmovant “must do more than simply show that there is some metaphysical doubt as to the material facts.” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986). It must show that there is sufficient evidence from which a reasonable factfinder could find in its favor. See Celotex, 477 U.S. at 322-23, 106 S.Ct. at 2552-53. This standard “mirrors the standard for a directed verdict under Federal Rule of Civil Procedure 50(a), which is that the trial judge must direct a verdict if, under the governing law, there can be but one reasonable conclusion.” Anderson, 477 U.S. at 250, 106 S.Ct. at 2511. In determining the sufficiency of the non-moving party’s evidence, all inferences to be drawn from underlying facts should be resolved in the favor of the nonmoving party, see Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356, but only such evidence as would be admissible at trial can be considered. See Wilson v. Clancy, 747 F.Supp. 1154, 1158 (D.Md.1990), aff'd, 940 F.2d 654 (4th Cir.1991).
III. POPULATION EQUALITY
A. The Legal Standards
The “one person, one vote” principle was first articulated by the Supreme Court in Gray v. Sanders, 372 U.S. 368, 83 S.Ct. 801, 9 L.Ed.2d 821 (1963). That principle, which is grounded in the Equal Protection Clause of the Fourteenth Amendment, prohibits the dilution of individual voting power by means of state districting plans that allocate legislative seats to districts of different populations. In Reynolds v. Sims, 377 U.S. 533, 568, 84 S.Ct. 1362, 1385, 12 L.Ed.2d 506 (1964), the Court held that “the Equal Protection Clause requires that the seats in both houses of a bicameral state legislature must be apportioned on a population basis.” Requiring electoral districts to be of nearly equal population ensures that each person’s vote is given the same weight. Thus, in Reynolds, the Court required States to “make an honest and good faith effort to construct districts ... as nearly of equal population as is practicable.” Id. at 577, 84 S.Ct. at 1390.
Although the Supreme Court has held that absolute population equality should be the paramount objective in plans allocating congressional districts, see Karcher v. Daggett, 462 U.S. 725, 732-33, 103 S.Ct. 2653, 2659-60, 77 L.Ed.2d 133 (1983), the Court affords more flexibility to States in formulating districting plans for state legislative seats by requiring only “substantial” population equality. See Gaffney v. Cummings, 412 U.S. 735, 748, 93 S.Ct. 2321, 2329, 37 L.Ed.2d 298 (1973). This slightly relaxed requirement for state redistricting plans recognizes that minor deviations from absolute population equality may be necessary to permit the States to pursue other legitimate state policies. See Reynolds, 377 U.S. at 577-81, 84 *1031S.Ct. at 1389-92; Mahan v. Howell, 410 U.S. 315, 321-22, 93 S.Ct. 979, 983-84, 35 L.Ed.2d 320 (1973). The Court has specifically recognized a number of state policies that justify minor deviations from absolute population equality. In the Karcher decision, a congressional redistricting case, the Court stated: “Any number of consistently applied legislative policies might justify some variance, including, for instance, making districts compact, respecting municipal boundaries, preserving the cores of prior districts, and avoiding contests between incumbent Representatives.” 462 U.S. at 740, 103 S.Ct. at 2663. Given the more stringent population equality standard in congressional redistricting, it is clear that any state policies that are recognized by the Court as sufficient to depart from absolute equality should also be sufficient to depart from the less stringent requirement of “substantial” population equality. See,- e.g., Legislative Redistricting Cases, 331 Md. at 594, 629 A.2d at 656 (applying Karcher’s list of legitimate state policies to the Maryland state legislative redistricting plan being challenged here).
1. Maximum Deviations Below Ten Percent. Because the promotion of these other important state policies will often, in combination, prevent States from attaining absolute population equality in a districting plan, the Supreme Court has established that “minor deviations” from mathematical population equality, alone, are insufficient to establish a prima facie case of invidious discrimination. Voinovich v. Quitter,- — U.S. -,-, 113 S.Ct. 1149, 1159, 122 L.Ed.2d 500 (1993); see also Swann v. Adams, 385 U.S. 440, 444, 87 S.Ct. 569, 572, 17 L.Ed.2d 501 (1967) (“De minimis deviations are unavoidable .... ”). Redistricting plans that have a maximum population deviation under ten percent fall within this category of minor deviations. Voinovich, — U.S. at-, 113 S.Ct. at 1159 (quoting Brown v. Thomson, 462 U.S. 835, 842, 103 S.Ct. 2690, 2696, 77 L.Ed.2d 214 (1983)). Thus, a redistricting plan with a maximum deviation below ten percent is prima facie constitutional and there is no burden on the State to justify that deviation. See Holloway v. Heckler, 817 F.Supp. 617, 623 (S.D.W.Va.1992) (three-judge court), aff'd mem., — U.S.-, 113 S.Ct. 1378, 122 L.Ed.2d 754 (1993); Fund for Accurate and Informed Representation, Inc. v. Weprin, 796 F.Supp. 662, 668 (N.D.N.Y.) (three-judge court), aff'd mem., — U.S. -, 113 S.Ct. 650, 121 L.Ed.2d 577 (1992); Gorin v. Karpan, 788 F.Supp. 1199, 1201 (D.Wyo.1992) (three-judge court); Cosner v. Dalton, 522 F.Supp. 350, 357 n. 11 (E.D.Va. 1981) (three-judge court).
The defendants argue that the “ten percent rule” essentially forecloses all challenges to redistricting plans in which the maximum deviation is below that percentage. The defendants support this position with the fact that no state districting plan with a maximum deviation below ten percent has ever been struck down by a court on the basis of a population equality violation. See Defs.’ Mem. in Supp. of Mot. for Summ.J. at 4-5. Indeed, the language of a few recent decisions could be read to support this position. The Weprin decision, for example, might be read to indicate that plaintiffs who fail to demonstrate a maximum population deviation in excess of ten percent are simply unable to establish a prima facie case of unconstitutional vote dilution. See Weprin, 796 F.Supp. at 668 (“[Ajbsent credible evidence that the maximum deviation exceeds 10 percent, plaintiffs fail to establish a prima facie case of discrimination under [the one person, one vote] principle sufficient to warrant further analysis by this Court.”); see also Legislative Redistricting Cases, 331 Md. at 597, 629 A.2d at 657 (“As long as the population disparities between legislative districts adhere to the requirements of Reynolds and its progeny — i.e. as long as maximum deviations are under 10% — disparities in the number of representatives from the various regions or political subdivisions in the State are prima facie immaterial.”).
The plaintiff, on the other hand, argues that there is a distinction between prima facie constitutional and per se constitutional. MFR concedes that if the maximum deviation of a redistricting plan is under ten percent, the State has no burden to justify that deviation. See Pl.’s Mem. in Opp’n to Defs.’ Mot. for Summ.J. at 4 [hereinafter “Opp’n Mem.”]. However, MFR argues that it could successfully challenge a plan with a maxi*1032mum deviation below ten percent if it can prove that the “minor deviation” in the plan does not result from the promotion of other' legitimate state policies, but rather from an impermissible or irrational purpose. See PL’s Supplemental Letter of Nov. 24,1993, in Opp’n to Defs.’ Mot. for Summ.J. at 2-3.
The Court concludes that a plan with a maximum deviation below ten percent could still be successfully challenged, with appropriate proof, for several reasons. First, although there is some language in several decisions indicating that plans with less than a ten percent deviation are essentially per se constitutional, none of those decisions expressly so states, and it appears that the plaintiffs in those cases did not raise any arguments that the minor deviation was the result of an unconstitutional or irrational purpose. See, e.g., Weprin, 796 F.Supp. at 668 (no claim of unconstitutional or irrational state policy). In fact, several decisions have expressly stated that redistricting plans with a minor deviation could be challenged with such a showing. See, e.g., Legislative Redistricting Cases, 331 Md. at 597, 629 A.2d at 657 (“Possibly, there may be room under Reynolds and its progeny for a plaintiff to overcome the 10% rule,’ if the plaintiff can present compelling evidence that the drafters of the plan ignored all the legitimate reasons for population disparities and created the deviations solely to benefit certain regions at the expense of others.”); Licht v. Quattrocchi, 449 A.2d 887, 887 (R.I.1982) (finding a maximum deviation of five percent violated the one person, one vote requirement when the deviation “negate[d] the effects of reapportionment” by enhancing the representation of two cities). Finally, if the defendants’ argument were correct, individuals whose votes are systematically diluted statewide solely to promote an unconstitutional or irrational state policy,8 would be without a remedy simply because the State was able to contain the maximum deviation of its plan within ten percent.
For these reasons, this Court holds that a plaintiff could, with appropriate proof, successfully challenge a redistricting plan with a maximum deviation below ten percent. To prevail, though, the plaintiffs have the burden of showing that the “minor” deviation in the plan results solely from the promotion of an unconstitutional or irrational state policy. Thus, the plaintiff must demonstrate, just as the defendants must demonstrate when the State has the burden of proving that the plan is constitutional, see infra, that the asserted unconstitutional or irrational state policy is the actual reason for the deviation. See Karcher, 462 U.S. at 740-44, 103 S.Ct. at 2663-67. In addition, the plaintiff must prove that the minor population deviation is not caused by the promotion of legitimate state policies.9
2. Maximum Deviations Above Ten Percent. If the maximum deviation of a districting plan exceeds ten percent, then the plan is prima facie unconstitutional and the larger disparities in population must be justified by the. State Voinovich, — U.S. at -, 113 S.Ct. at 1159; Brown, 462 U.S. at 842-43, 103 S.Ct. at 2695-96. The State has the burden of demonstrating that the plan may reasonably be said to advance a rational state policy, such as those listed in the Karcher decision, and, if so, that “the population disparities among the districts that have resulted from the pursuit of this plan [do not] *1033exceed constitutional limits.” Voinovich, — U.S. at -, 113 S.Ct. at 1159 (quoting Brown, 462 U.S. at 843, 103 S.Ct. at 2696 (quoting Mahan, 410 U.S. at 328, 93 S.Ct. at 987)) (internal quotation- marks omitted). Thus, if the deviation of the districting plan exceeds ten percent, the State has the burden of justifying that deviation. The defendants have the burden of expressly identifying the specific rational state policies the advancement of which has caused the deviation. See Karcher, 462 U.S. at 740-44, 103 S.Ct. at 2663-67. In addition, the State has the burden of demonstrating that the deviation does not exceed constitutional limits.
MFR, claiming to state the standard enunciated in Voinovich, argues that the State must demonstrate that the deviation is required in order to advance a rational state policy. See Opp’n Mem. at 6. This is simply not the standard that has been established for evaluating deviations in state legislative redistrieting plans. Rather, MFR has stated the test for congressional redistrieting cases as set out in Karcher, 462 U.S. at 741, 103 S.Ct. at 2664. The Karcher decision is applicable to state legislative redistrieting cases to the extent that it expressly recognizes several state policies that might justify deviations from population equality. The requirement in Karcher that the asserted state policy actually be the reason for the deviation is equally applicable to state redistrieting cases. However, it is well established that the one person, one vote standard for congressional redistrieting is far more demanding than the standard for state legislative redistrieting plans. See Mahan, 410 U.S. at 321-23, 93 S.Ct. at 983-85; Reynolds, 377 U.S. at 578, 84 S.Ct. at 1390. As the defendants correctly point out, the appropriate test to be applied in state legislative redistricting cases is more akin to a rational basis test. “ ‘So long as the divergences from a strict population standard are based on legitimate considerations incident to the effectuation of a rational state policy, some deviations from the equal-population principle are constitutionally permissible with respect to' the apportionment of seats in either or both of the two houses of a bicameral state legislature.’ ” Mahan, 410 U.S. at 325, 93 S.Ct. at 985 (quoting Reynolds, 377 U.S. at 579, 84 S.Ct. at 1391) (emphasis added).
Before discussing whether the State’s plan violates the population-equality requirement of the Equal Protection Clause, the arguments of both parties raise the preliminary question of whether the senate and delegate plans should be analyzed as an entire plan or separately. In their motion for summary judgment, the defendants analyze the constitutionality of the plan by discussing the senate and delegate portions of the plan separately. See Defs.’ Mem. in Supp. of Mot. for Summ.J. at 4-8. MFR asserts that this division is meaningless and that if either portion of the plan is found to be unconstitutional, then the entire plan is unconstitutional. See Opp’n Mem. at 2, 6. The Supreme Court has expressly stated that courts should be reluctant to strike down an entire apportionment plan when the constitutional infirmity could be cured by lesser means. See Davis v. Bandemer, 478 U.S. 109, 137 n. 16, 106 S.Ct. 2797, 2813 n. 16, 92 L.Ed.2d 85 (1986) (plurality opinion); Whitcomb v. Chavis, 403 U.S. 124, 160-61, 91 S.Ct. 1858, 1878-79, 29 L.Ed.2d 363 (1971). In this case, the delegate portion of the plan utilizes multi-member, two-member, and single-member delegate districts that are nested within specific senate districts. Thus, if the Court finds a constitutional defect in the delegate portion of the plan only, it could be remedied by redrawing the delegate district lines within the plan’s acceptable senate districts. A separate analysis of the senate and delegate portions of the plan is therefore appropriate.
B. The Senate Plan
.It is undisputed that the maximum variance between senate districts in the State’s redistrieting plan is 9.84%. Thus, MFR cannot, based solely on population variation, establish a prima facie case of unconstitutional discrimination, and the State need not justify the plan’s minor deviation. See Voinovich, — U.S. at ——, 113 S.Ct. at 1159; Brown, 462 U.S. at 842-43, 103 S.Ct. at 2695-96. The burden remains on MFR to prove that the deviation is the result of an unconstitutional or irrational state purpose rather than one or more of the other state *1034policies already recognized as legitimate by the Supreme Court.
MFR concedes that it has the burden to prove that the plan violates the one person, one vote requirement, but argues that it has advanced sufficient evidence of discrimination not only to withstand the State’s motion for summary judgment, but also to prevail on its own cross-motion for summary judgment. In support of this position, MFR advances two arguments. First, MFR argues that the plan is invalid because the GRAC, from the inception of the redistricting process, had the goal of containing the maximum population deviation within plus or minus five percent of the ideal population, rather than the objective of absolute population equality. See Opp’n Mem. at 7, 11. MFR’s second argument is that the redistricting plan is invalid because the GRAC had the unconstitutional purpose of providing Baltimore City with control of eight senatorial districts even though -the population of the City, alone, could not support that many districts. See id. at 9-11.
1. Utilization of a Population Window. In support of its first argument, MFR refers to a document titled “Legal Standards for Plan Development,” as well as to the deposition testimony of three members of the Committee. See Opp’n Mem. at 7, 10-12. The “Legal Standards” document provided, in part: “Deviations from the ideal Senate district or legislative subdistrict population should not exceed + 5% or - 5% (except as may be necessitated by other constitutional requirements).” See Legal Standards for Plan Development at II.A.2.(a). There is also no dispute that the members of the Committee all had the goal of containing the maximum population deviation of the plan to within ten percent. See, e.g., Aro Dep. at 138.
It is therefore undisputed that the GRAC had the objective at the inception of the redistricting process of a ten percent maximum population deviation rather than absolute population equality. Far from being an unconstitutional or irrational purpose, however, this objective merely recognized the flexibility that the State had in order to accommodate other legitimate state policies. The Supreme Court has expressly provided States with a degree of flexibility, ie., a ten percent population deviation, in formulating a state legislative districting plan because of the legitimate state interests that might cause deviations from absolute population equality. See Voinovich, — U.S. at-, 113 S.Ct. at 1159; Reynolds, 377 U.S. at 577-81, 84 S.Ct. at 1389-92. A population deviation under ten percent is considered “minor” and a State’s districting plan with only a minor deviation is prima facie constitutional, or, in other words, has achieved substantial population equality in conformity with the Equal Protection Clause. See Mahan, 410 U.S. at 322-25, 93 S.Ct. at 984-85. Thus, an express objective of containing the population deviation of the plan to within ten percent demonstrates nothing more than the objective of crafting a plan with constitutional population equality.10 An argument similar to MFR’s was rejected by a three-judge district court in Farnum v. Burns, 561 F.Supp. 83 (D.R.I.1983), which held that, because the purpose of the ten percent rule was to give flexibility to incorporate other legitimate state policies into a redistricting plan, “it does not violate the one-person, one-vote principle merely because a population window was used in developing that plan.” Id. at 93. This Court quite agrees.
Finding that the use of a plus or minus five percent population window is not an illegitimate state purpose or objective, the Court also finds that the use of that window supports no inference that the population deviation was not due to the promotion of legitimate state policies. In fact, if the use of the population window supports any inference at all, it is that the Committee had the intention, from the inception of the redistricting process, to incorporate those other court-*1035approved state policies into the plan.11
2. Discrimination in Favor of Baltimore City. MFP next argues that the senate plan is unconstitutional because one of the goals of the GRAC, from the beginning of the redis-trieting process, was to provide Baltimore City with eight senate seats, even though the population of the City, by itself, could not support eight districts and remain within the ten percent population deviation goal. There is clearly sufficient evidence to demonstrate that this was at least one of the Committee’s objectives. See Brown Dep. at 26; Glasgow Dep. at 65. It is also clear that in order to provide Baltimore with eight senate districts it was necessary to cross subdivision boundaries into Baltimore County repeatedly. See Glasgow Dep. at 76-78.
As the Court of Appeals of Maryland noted in its review of this redistricting plan, whether maximizing the legislative representation of a region with waning political influence but urgent political needs should be recognized under Reynolds as a rational state policy is an interesting question. Legislative Redistricting Cases, 331 Md. at 598 n. 19, 629 A.2d at 658 n. 19. It is clear that, as applied to Baltimore City, this was an objective of the GRAC. But, even assuming arguendo that this objective was unconstitutional,12 MFR is completely unable to satisfy its burden of proof of causation and the defendants are entitled to summary judgment.
Throughout its brief, the plaintiff has argued that the Karcher decision requires the State, when it has the burden of proving the plan’s constitutionality, to demonstrate that the asserted state policy is the actual reason for the plan’s deviation. See, e.g., Opp’n Mem. at 12 (“The burden on the State is to prove that the actual reason for the deviation was a Court-approved, rational State poli*1036cy.”) (citing Karcher, 462 U.S. at 741, 103 S.Ct. at 2664). Applying this same standard to the plaintiff when it has the burden of proving that the plan is unconstitutional makes perfect sense. In this context, MFR has no proof whatsoever that the incorporation of this objective into the redistricting plan was the cause of the plan’s minor deviation between the populations of senate districts across the State.
The Court is unable to locate anywhere in the record any evidence, other than concluso-ry assertions, that the deviation in the senate plan was caused by this objective. The only evidence that is arguably relevant to this issue is deposition testimony of Karl Aro, Deputy Director of the State Department of Legislative Reference, and of Michel A. Let-tre, Assistant Director of the Office of Planning, that plaintiff cites in its Opposition Memorandum. That testimony indicates that a senate map could have been drawn that would have had equal district populations within fifty to one hundred people. See Opp’n Mem. at 10-12. However, this does not demonstrate that the deviation in the plan was caused by the decision to provide Baltimore City with control over eight senatorial districts. It is not even clear from that testimony whether such a map could have been drawn incorporating the numerous court-approved legitimate state policies. For these reasons, the plaintiff is simply unable to prove the necessary causation between the asserted unconstitutional state objective and the deviation in the senate plan.
In addition, MFR cannot prove that the deviation was not caused by the promotion of court-approved state policies. Rather, there is evidence demonstrating that those factors were in fact considered. For example, as stated above, the “Legal Standards” document identifies several recognized state policies such as compactness, respecting political boundaries, preserving the cores of prior districts, and avoiding contests between incumbent representatives. See Karcher, 462 U.S. at 740, 103 S.Ct. at 2663. Also, in his deposition, Michel Lettre testified that the Committee had the goals of “compactness, contiguity, adherence to political subdivision lines, communities of interest and the like.” Lettre Dep. at 158. The GRAC members whose testimony was cited by the plaintiff to demonstrate that one of the objectives of the Committee was to provide Baltimore City with control over eight senate seats indicated that they also relied on other legitimate policies. Donna Felling indicated that the GRAC attempted to maintain the integrity of districts, to avoid contests between incumbents, and to comply with the Voting Rights Act. See Felling Dep. at 60. Chairman Benjamin Brown stated that the GRAC considered contiguity, compliance with the Voting Rights Act, and avoidance of contests between incumbents. See Brown Dep. at 56-59. Thus, even looking at the evidence in the light most favorable to the plaintiff, see Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356, MFR is simply unable to demonstrate that the minor deviation in the Senate plan is not due to the promotion of court-approved legitimate state policies. Accordingly, we grant the State’s motion for summary judgment on the MFR’s population-equality claim regarding the senate portion of the plan and deny MFR’s cross-motion for summary judgment.
C. The House of Delegates Plan
It is undisputed that the maximum variance between delegate districts in the State’s redistricting plan is 10.67%. A population deviation in excess of ten percent is prima facie unconstitutional and the burden is on the State to demonstrate that the plan may reasonably be said to advance rational state policies, such as those listed in the Karcher decision, and that the population disparities among districts that have resulted from the pursuit of this plan do not exceed constitutional limits. See Voinovich, — U.S. at -, 113 S.Ct. at 1159 (quoting Brown, 462 U.S. at 842-43, 103 S.Ct. at 2695-96).
There are four delegate districts that deviate more than five percent from the ideal district population: District 2A in Washington County and District 35B in Cecil County are relatively small; District 27A in Prince George’s County and District 29C in Calvert and St. Mary’s Counties are relatively large. The State must first demonstrate *1037that the variations in these districts are due to the advancement of rational state policies. See Voinovich, — U.S. at-, 113 S.Ct. at 1159. In Reynolds and in subsequent decisions, several state policies have been recognized that might justify deviation from absolute population equality. These include respecting the boundaries of political subdivisions, Reynolds, 377 U.S. at 578, 84 S.Ct. at 1390, preserving the cores of prior districts, avoiding contests between incumbents, Karcher, 462 U.S. at 740, 103 S.Ct. at 2663, and due regard for natural boundaries, In re Legislative Districting, 299 Md. 658, 681-82, 475 A.2d 428, 440, appeal dismissed, 459 U.S. 962, 103 S.Ct. 286, 74 L.Ed.2d 272 (1982).
In support of their motion, the defendants have submitted the affidavit of Ronald M. Kreitner, Director of the Maryland Office of Planning. (That office provided the Committee with staff to assist it in preparing the congressional and state legislative redistricting plans. See Kreitner Aff. at ¶ 3.) In his affidavit, Kreitner described in detail the state policies that required the populations of those four delegate districts to deviate more than five percent from the ideal district size. Id. at ¶¶ 18-21. Kreitner stated that District 2A has a .deviation of — 5.52% in order to preserve the boundaries of three political subdivisions: the City of Hagerstown and Washington and Frederick Counties. Id. at ¶ 18. District 27A has a deviation of + 5.09% that Kreitner states is due to preserving county lines and the cores of prior districts where possible. Id. at ¶ 19. The deviation of + 5.24% in District 29C was necessary to avoid splitting precincts and to respect natural boundaries. Id. at ¶ 20. Finally, District 35B has a deviation of — 5.15% because it gives due regard to natural boundaries and preserves the cores of districts. Id. at ¶ 21. As noted above, all of these policies have been recognized by prior judicial decisions.
MFR simply has advanced no evidence to refute that these deviations were based on legitimate considerations incident to the ef-fectuation of rational state policies. See Reynolds, 377 U.S. at 579, 84 S.Ct. at 1390-91. Rather, it attempts to attack the sufficiency of Kreitner’s affidavit. MFR asserts that Kreitner took no part in the creation of the plan and that his statements are merely after-the-fact rationalizations. See Opp’n Mem. at 12-13. However, it is undisputed that the Maryland Office of Planning, of which Kreitner was the Director, supplied staffing to the Committee. MFR has presented no evidence in a form that may be considered in resolving a motion for summary judgment, see Wilson, 747 F.Supp. at 1158, that demonstrates that Kreitner lacks the knowledge to which he attests. For this reason, the Court finds that the defendants haye satisfied the first portion of the test enunciated in Reynolds and have demonstrated that the deviations in the four districts in question were based on legitimate considerations incident to the effectuation of rational state policies.
The defendants must next prove that the population disparities among the districts that have resulted from the pursuit of the plan do not exceed constitutional limits. See Voinovich, — U.S. at -, 113 S.Ct. at 1159 (quoting Brown, 462 U.S. at 843, 103 S.Ct. at 2696 (quoting Mahan, 410 U.S. at 328, 93 S.Ct. at 987)). The Court finds that this requirement is easily met here. Although the total maximum deviation of the plan exceeds ten percent, the Supreme Court has previously approved state districting plans that have had higher deviations and have been based on fewer legitimate state policies. Perhaps the best example is the Mahan decision, in which the Court validated a Virginia districting plan with a maximum deviation in excess of sixteen percent and justified only by the policy of respecting county and city boundaries. See 410 U.S. at 329, 93 S.Ct. at 987. In the present case there were numerous state policies that contributed to a maximum deviation on the delegate side just barely in excess of ten percent. For this reason, the Court finds that the defendants have met their burden under Reynolds and its progeny. Thus, we grant their motion for summary judgment regarding MFR’s population-equality claim for the delegate portion of the state plan. MFR’s cross-motion for summary judgment is, therefore, denied.
*1038IV. POLITICAL GERRYMANDERING
The plaintiff must satisfy a two-part test to prevail on a claim of unconstitutional political gerrymandering. In Davis v. Ban-demer, the Supreme Court confirmed by plurality opinion that a plaintiff must “prove both intentional discrimination against an identifiable political group and an actual discriminatory effect on that group.” Bandemer, 478 U.S. at 127, 106 S.Ct. at 2808. Although MFR could withstand the defendants’ summary judgment motion on the issue of discriminatory intent, it is clear that it cannot produce sufficient evidence to satisfy the demanding test established in Bandemer to demonstrate discriminatory effect. Thus, we grant the defendants’ motion for summary judgment on MFR’s political gerrymandering claim, for reasons that follow.
A. Intentional Discrimination
Although expressly stating that discriminatory intent must be proved, the plurality opinion in Bandemer recognized that, “[a]s long as redistricting is done by a legislature, it should not be very difficult to prove that the likely political consequences of the reapportionment were intended.” 478 U.S. at 129, 106 S.Ct. at 2809; see id. at 129 n. 11, 106 S.Ct. at 2809 n. 11. Subsequent to Ban-demer, several lower federal courts have acknowledged that the standard for establishing discriminatory intent is relatively undemanding. See, e.g., Republican Party v. Martin, 980 F.2d 943, 955 (4th Cir.1992) (“The intent standard set forth in the Ban-demer plurality opinion is easily met....”), reh’g denied, 991 F.2d 1202 (4th Cir.), and cert. denied, — U.S.-, 114 S.Ct. 93, 126 L.Ed.2d 60 (1993); Badham v. Eu, 694 F.Supp. 664, 670 (N.D.Cal.1988) (stating that Bandemer’s intent requirement is easily satisfied, and focusing instead on allegations of discriminatory effects), aff'd mem., 488 U.S. 1024, 109 S.Ct. 829, 102 L.Ed.2d 962 (1989).
The Supreme Court had recognized previously that any redistricting process is inherently political and that “[district lines are rarely neutral phenomena.” Gaffney v. Cummings, 412 U.S. at 753, 93 S.Ct. at 2331.
They can well determine what district will be predominantly Democratic or predominantly Republican, or make a close race likely. Redistricting may pit incumbents against one another or make very difficult the election of the most experienced legislator. The reality is that districting inevitably has and is intended to have substantial political consequences.
Id. In fact, the Court in Gaffney went on to state that it is to be expected that those who formulate a redistricting plan seek to achieve the political ends “of the State, its constituents, and its officeholders.” Id. at 754, 93 S.Ct. at 2332.
Drawing all inferences from the underlying facts in favor of the nonmóving party, see Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356, MFR has advanced sufficient evidence, under the weak showing required by Bandemer, to defeat the defendants’ motion for summary judgment on the issue of discriminatory intent. Although several factors might be cited by the defendants to negate an inference of discriminatory intent,13 MFR has raised several facts that sufficiently support the inference that the plan was formulated with the intention of discriminating against Republicans and in favor of .Democrats. The evidence is undisputed that the GRAC was provided with a binder entitled “Democratic Voting Performance and Strength” which contained comprehensive information on Democratic voter distribution throughout the State and ranked each precinct by Democratic strength. See Lettre Second Dep. at 13. Michel Lettre stated in *1039his deposition that he prepared and presented a report to the Committee from which its members could easily determine Democratic strength and preference by district. See id. at 7, 26.
A top administrative aide to Senator Thomas V. “Mike” Miller, President of the Senate and a member of the Committee, was provided with a computer database (the “voter file”) that had the capability of showing any precinct’s voting history. See Landow Dep. at 19-21. Thomas Cowley, former executive director of the Maryland Democratic Party, stated that the voter file was created specifically to aid in redistricting and to give Democratic candidates a partisan advantage. See Cowley Dep. at 44. Although the voter file was created and utilized for the State’s congressional redistricting plan, it was not returned after completion of that, plan and no restrictions were placed on the Committee’s use of the file. See id. at 64-65.
Finally, both the House Minority Leader and the House Minority Whip testified at their depositions' that they had been told by House Democrats that the plan was expected to diminish the relatively small number of House seats already held by Republicans. See Sauerbrey Dep. at 56; Kittleman Dep. at 79.
All of these facts, when viewed in the light most favorable to MFR, see Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356, support a reasonable inference that the plan was intended to discriminate against Republican candidates. Because MFR has generated sufficient evidence from which a reasonable factfinder could infer a discriminatory intent, there is a genuine dispute as to a material fact on that issue. See id. at 586, 106 S.Ct. at 1355-56. However, MFR can defeat the defendants’ motion for summary judgment only if it is also able to generate a genuine dispute as to the second prong of the Ban-demer test, discriminatory effect.
B. Discriminatory Effect
In developing its test for establishing discriminatory effect in Bandemer, the plurality utilized previous opinions discussing claims for racial gerrymandering. See Bandemer, 478 U.S. at 131, 132 n. 13, 136-37, 106 S.Ct. at 2809, 2810 n. 13, 2812-13. Bandemer also acknowledged that the test for establishing statewide political gerrymandering, although formulated a bit differently, was essentially the same as the tests established for challenges to individual multimember districts. “[T]he question is whether a particular group has been unconstitutionally denied its chance to effectively influence the political process.” Id. at 132-33, 106 S.Ct. at 2810. MFR must, therefore, produce evidence of either a continued frustration of the will of a majority of the voters or an effective denial to a minority of voters of the fair opportunity to influence the political process as a whole. Id. at 133, 106 S.Ct. at 2810. Because this is not a case in which it is alleged that the will of a majority of voters continues to be frustrated,14 MFR must be able to demonstrate that an identifiable minority of voters (i.e., Republicans) has been denied its chance to influence the political process effectively.
The- ability to influence the political process as a whole is not limited to the ability to win elections. See id. at 132, 106 S.Ct. at 2810. The mere fact that a particular redistricting plan makes it more difficult for a particular group to elect the representatives of its choice does not render the plan constitutionally infirm. Id. at 131, 106 S.Ct. at 2809. Even in “safe” districts in which the majority party consistently wins elections year after year, the group of individuals who voted for the losing candidate is presumed to have as much opportunity to influence the victorious candidate as other voters in the district. Id. at 132, 106 S.Ct. at 2810; Badham, 694 F.Supp. at 670-71. In the statewide gerrymandering context, the Bandemer plurality acknowledged that because of the nature of winner-take-all, district-based elections, disproportionate election results are likely to occur whenever there is even a narrow statewide preference for one party. See 478 U.S. at 130, 106 S.Ct. at 2809. Thus, MFR must be able to demonstrate more than *1040simply losing elections disproportionately on a statewide scale. A history of disproportionate results must appear “in conjunction with strong indicia of lack of political power and the denial of fair representation.” Id. at 139, 106 S.Ct. at 2814. MFR must prove that Republican voters have “essentially been shut out of the political process.” Id. Although not expressly so stating, Bandemer appeared to have concentrated on two areas of inquiry: participation in the electoral process and the responsiveness of elected officials.
1. Participation in Election Process. First, the districting plan should be examined to determine to what extent the minority group is able to participate in the election process. Id. at 133, 106 S.Ct. at 2811 (plurality opinion). Relying on racial gerrymandering decisions, Bandemer noted the importance of a minority group’s ability to participate “in the slating and nomination of candidates, their opportunity to register and vote, and hence their chance to directly influence the election returns and to secure the attention of the winning candidate.” Id. The plurality opinion acknowledged that this participatory approach might not be as helpful when analyzing a claim that a plan discriminates against a minority political party. It would indeed be difficult for members of a political party to demonstrate that they are excluded from participating in the affairs of their own party or from the processes by which candidates are nominated and elected. See id. at 137, 106 S.Ct. at 2812-13.
However, the electoral participation inquiry has been adapted by several lower courts, subsequent to Bandemer, to apply more appropriately to claims of political gerrymandering. In Republican Party v. Martin, the Fourth Circuit was confronted with a claim that North Carolina’s scheme for electing superior court judges discriminated against Republicans. Although the court recognized that the factors that the Supreme Court had articulated regarding racial gerrymandering did not apply particularly well to claims of political gerrymandering, the court found that the allegations in the complaint were sufficient to indicate that the North Carolina scheme inhibited the ability of Republicans to participate in the election process. See 980 F.2d at 956-58. The court noted that Republicans were discouraged from running because of the high likelihood of losing, and, in turn, campaign contributions were seriously diminished because contributors were less willing to donate money to candidates who were perceived to be almost certain losers. Id. at 957.
In Badham v. Eu, a three-judge district court dismissed a gerrymandering claim made by state Republicans because they were unable to demonstrate “that anyone has ever interfered with Republican registration, organizing, voting, fund-raising, or campaigning.” 694 F.Supp. at 670. The court noted that Republicans remained free to speak out on public issues and were not inhibited in any way from participating in the public debate on which our political system relies. Id. Finally, in Pope v. Blue, 809 F.Supp. 392 (W.D.N.C.) (three-judge court), aff'd mem., — U.S. -, 113 S.Ct. 30, 121 L.Ed.2d 3 (1992), the district court dismissed a claim by Republicans of political gerrymandering because the complaint failed to allege any disruption of the electoral activities listed in Badham. See Pope, 809 F.Supp. at 397 (quoting Badham, 694 F.Supp. at 670).
2. Responsiveness of Elected Officials. The other factor that the Bandemer plurality emphasized in determining whether a particular group has been unconstitutionally excluded from the political process was whether the victorious elected officials were responsive to the interests of the complaining minority. See Bandemer, 478 U.S. at 131-32, 106 S.Ct. at 2809-10. The Court’s prior decisions in the context of racial gerrymandering indicate that a court must find that the districting scheme is set up in such a way that the elected representatives do not need to be and are not in fact responsive to the interests of the minority. In White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973), for example, the Court noted that a white-dominated organization (the Dallas Committee for Responsible Government) effectively controlled the Democratic Party, enabling white candidates to win elections without any support from the black community. Id. at 766-67, 93 S.Ct. at 2339-40. In addi*1041tion, the Committee did not “exhibit good-faith concern for the political and other needs and aspirations of the Negro community.” Id. at 767, 93 S.Ct. at 2340 (describing the district court’s findings). However, in Ban-demer, the plurality expressly stated that courts should not presume, without specific supporting evidence, that those who are elected will ignore the interests of underrepresented groups. See 478 U.S. at 132, 106 S.Ct. at 2810.
In its motion for summary judgment, the defendants argue that MFR is simply unable to identify any facts that are sufficient to satisfy the demanding test for discriminatory effect established by the Bandemer decision. MFR relies almost completely on the deposition testimony of Delegate Ellen Sauerbrey, House Minority Leader, and Delegate Robert Kittleman, House Minority Whip, arguing that their deposition testimony, if taken in the most favorable light, could demonstrate to a reasonable factfinder that without a different redistricting plan Republicans are, and will continue to be, “shut out” of the political process. MFR asserts that this testimony indicates that, at the statewide level, the Maryland General Assembly is essentially a one-party system that is unresponsive to the interests of the minority Republicans and to those whom they represent. See Opp’n Mem. at 20-22.
The facts on which MFR relies, even if assumed to be true, simply do not satisfy, as a matter of law, the test for discriminatory effect established in Bandemer. MFR has indisputably demonstrated that Republicans have been, and will continue to be, underrepresented in the Maryland General Assembly. Democrats are consistently elected to the House and Senate in numbers out of proportion to their statewide vote totals or registration numbers. While Republicans constitute around thirty percent of all registered voters, they hold somewhat less than twenty percent of the total seats in the House and Senate.
A mere showing of disproportionate election results is insufficient, however. The Supreme Court has expressly held that disproportionate statewide election results simply do not render a legislative districting plan unconstitutional. Such results are simply inherent in winner-take-all, district-based elections in which even a narrow statewide preference for a party can give it an overwhelming majority of seats in the state legislature. See Bandemer, 478 U.S. at 130, 106 S.Ct. at 2809 (plurality opinion); Whitcomb, 403 U.S. at 160, 91 S.Ct. at 1878. MFR must also demonstrate that this history of disproportionate election results appears “in conjunction with strong indicia of lack of political power and the denial of fair representation.” See Bandemer, 478 U.S. at 139, 106 S.Ct. at 2814 (plurality opinion).
To determine whether the redistricting plan effectively denies Republicans the ability to influence the political process, this Court must apply the inquiries discussed in Bandemer and examine the effect, if any, the plan will likely have on the ability of Republicans to participate in the electoral process. In addition, this Court should analyze the extent to which Republicans play a role in the legislative process within the General Assembly and whether the majority party “entirely ignore[s]” their interests. Id. at 132, 106 S.Ct. at 2810.
MFR has advanced no evidence that suggests that the redistricting plan in any way affects the ability of Republicans to participate effectively in the electoral process. There is no evidence that the plan and the alleged “advantage” that it affords to Democrats across the State has prevented or will prevent Republicans from raising money to support their campaigns. See Republican Party v. Martin, 980 F.2d at 957. There is no evidence that because of the structure of the plan, Republicans will be discouraged from running for state office, or that the plan will discourage voters from registering as Republicans. Compare id. (relying on fact that the statewide system of electing local judges was discouraging prospective Republican candidates from running due to poor election chances). Finally, there is no evidence to support any claim that Republicans, due to the structure of the plan, will be unable to participate fully in the “ ‘uninhibited, robust, and wide-open’ ” public debate over issues of concern to voters. See Bad-ham, 694 F.Supp. at 670 (quoting New York Times Co. v. Sullivan, 376 U.S. 254, 270, 84 *1042S.Ct. 710, 721, 11 L.Ed.2d 686 (1964)). In fact, the testimony of Delegates Sauerbrey and Kittleman indicates that the increased press coverage of Republican viewpoints in Annapolis has enabled the public to hear more than one side of an issue. See Kittle-man Dep. at 71; Sauerbrey Dep. at 43.
The fact that Republicans are elected in the numbers that they are actually supports the inference that they have not been excluded from effective participation in the electoral process. With plurality (winner-take-all) district-based elections, one can expect a party with even a narrow majority statewide to win a much larger proportion of seats than its proportion of the statewide vote. See Lichtman Aff. at ¶ 9. This phenomenon, known in the literature as the “swing ratio,” has been expressly acknowledged by the Supreme Court, which has refused to invalidate such winner-take-all elections as unconstitutional. See Bandemer, 478 U.S. at 130, 106 S.Ct. at 2809 (plurality opinion). While Democrats enjoy a two-to-one advantage over Republicans in the number of registered voters, the “differential between votes and seats represents only a modest swing ratio or multiplier effect compared to that of most states, especially given the overwhelmingly Democratic composition of the House and Senate votes.” Lichtman Aff. at ¶ 33. One would expect Republicans to be elected in far fewer numbers if they were effectively excluded from participating in the electoral process.
MFR argues that the deposition testimony of Delegates Sauerbrey and Kittleman demonstrates that Republicans have been sufficiently shut out of the political process to satisfy the effects requirement of Bandemer because it establishes that the Democratic leadership in the General Assembly is not responsive to their minority interests. Ban-demer expressly stated that there is no presumption that those who are elected by the majority will disregard the interests of disproportionately underrepresented groups. Bandemer, 478 U.S. at 132, 106 S.Ct. at 2810 (plurality opinion). The testimony that MFR advances establishes, at best, that the Republicans are indeed a minority party in the legislature of a predominantly Democratic State. There is no dispute that much of the business of the House of Delegates is conducted by its five standing committees and that the membership on those committees is proportional to party affiliation. There is no dispute that the position of Speaker of the House carries with it many powers, including the power to appoint members to committees and the power to control debate and voting on the floor of the House. It is only political reality that the Democratic Party, the political party with an overwhelming majority of registered voters and legislative seats, utilizes its power to control these processes in order to achieve its political agenda. Ban-demer requires more, however. MFR must be able to demonstrate that Democrats, in their control of the legislative processes, are able to, and do in fact, ignore Republican interests entirely. See id.
The deposition testimony advanced by MFR demonstrates that this is simply not the case. The Republican minority is the most powerful minority within the legislature and the Democratic leadership is responsive to its interests. Delegate Kittleman testified that the Republicans are the most influential minority in the House:
The leadership now has to contend with a vocal and active caucus.... [A]ny person leading any organization would as soon not have dissidence, but it is there now and it has to be recognized and it has to be considered in decisions.... [Wjhen we had sixteen members, I’d say the Republican caucus was about the third ranking body ... of minorities in the House; ... the black caucus and the women’s caucus were more effective than the Republican caucus. Now the Republican caucus is listened to more than the other two, after the Democrats, the number one dissenting group.
Kittleman Dep. at 36. Based on this status, the majority leadership frequently contacts and consults the minority leadership to determine its position. See Kittleman Dep. at 37; Sauerbrey Dep. at 74. Republicans have been able to build coalitions with like-minded Democrats and other minorities to pass legislation. See Kittleman Dep. at 72; Sauerbrey Dep. at 61. Numerous bills that were sponsored by Republicans are passed by the Gen*1043eral Assembly each year. See Defs.’ Resp. to Tiches’ Interrogs. to SABEL #25.
It is clear from this testimony that due to increased public awareness of Republican viewpoints and the ability of the Republican caucus to form voting coalitions with other minorities or like-minded Democrats, the Democratic leadership is simply unable to ignore Republican interests. It is certainly true that Republicans are often outvoted both in legislative committees and on the floor. This does not mean, however, that they have been unconstitutionally shut out of the political process as a whole. There is a difference between having a voice and being listened to. Bandemer requires that Republicans have no voice in the political process, which, as demonstrated by MFR’s own witnesses, they clearly do. When, to what extent, and on which issues they are listened to are questions that are quite properly resolved by the political process, not by the courts.
Finally, even if it is assumed that MFR can demonstrate that Republicans are shut out of the political process by the Democratic “super-majority” in the General Assembly, they have not demonstrated that any other alternative redistricting plan, even a plan establishing proportional representation, would significantly diminish Democratic control. Democrats constitute more than two-thirds of the total (two-party) registration in the State. See Lichtman Aff. at ¶ 33 n. 9. Although an increase in the number of legislative seats held by Republicans might make it possible for that party to affect the Democrats’ ability to pass constitutional amendments, to suspend rules, and to uphold a gubernatorial veto, even Delegate Sauerbrey concedes that proportional representation would not assure such powers to Republicans and that coalitions with dissident Democrats would probably still be required. See Sauer-brey Dep. at 53.
Although MFR has advanced sufficient evidence to create a factual dispute over discriminatory intent, it cannot, assuming all of its facts as true and drawing all inferences in its favor, satisfy the demanding test for discriminatory effect. There is no evidence that the plan has interfered with the ability of Republicans to participate in the electoral process. The evidence also demonstrates that the Democratic leadership in the General Assembly is, and must be, responsive to Republican interests. The State’s motion for summary judgment will be granted because MFR is unable, as a matter of law, to satisfy the two-part test established in Bandemer.
V. THE VOTING RIGHTS ACT CLAIMS
The plaintiffs in Civil Action No. S-92-1409 assert two Voting Rights Act claims. First, they allege that the State’s legislative redistricting plan “packs” and “cracks” minority voters statewide. Second, they allege that on the Eastern Shore the plan “submerges” a geographically compact and politically cohesive black community in two majority-white, at-large delegate districts.15
A. The Legal Standards
1. Section 2 Minority Vote Dilution Claims. Congress enacted § 2 of the Voting Rights Act of 1965, 79 Stat. 437, as amended, 42 U.S.C. § 1973, to help effectuate the Fifteenth Amendment’s guarantee that no citizen’s right to vote shall “be denied or abridged ... on account of race, color, or previous condition of servitude.” U.S. Const. amend. XV, § 1; see also Voinovich, — U.S. at-, 113 S.Ct. at 1154-55. Section 2 of the Voting Rights Act, as amended, 96 Stat. 134 (1982), provides:
SEC 2. (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 [because the citizen is a member of a language minority group], as provided in subsection (b) of this section.
*1044(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.
As the Supreme Court explained in Chisom v. Roemer, 501 U.S. 380, 111 S.Ct. 2354, 115 L.Ed.2d 348 (1991), “Section 2(a) adopts a results test, thus providing that proof of discriminatory intent is [unnecessary] to establish any violation of the section. Section 2(b) provides guidance about how the results test is to be applied.” Id., at-, 111 S.Ct. at 2364 (emphasis added). Section 2, taken as a whole, “prohibits any practice or procedure that, ‘interact[ing] with social and historical conditions,’ impairs the ability of a protected class to elect its candidate of choice on an equal basis with other voters.” Voinovick, — U.S. at -, 113 S.Ct. at 1155 (quoting Thornburg v. Gingles, 478 U.S. 30, 47, 106 S.Ct. 2752, 2764, 92 L.Ed.2d 25 (1986)).
Minority vote dilution can occur in either of two kinds of districting structures: (1) in an at-large, multimember district system, where the white majority consistently votes together as a bloc to “submerge” minority voters and to defeat their preferred candidates; or (2) in a single-member district system, where the minority voters are (a) “cracked,” i.e., fragmented and dispersed into two or more districts where they constitute an ineffective minority of the electorate, or (b) “packed,” ie., concentrated into a district where they constitute an excessive majority, or (c) both cracked and packed. See Voinovich, — U.S. at -, 113 S.Ct. at 1155; Gingles, 478 U.S. at 46 n. 11, 106 S.Ct. at 2764 n. 11.
£ The Gingles Threshold Inquiry. In Thornburg v. Gingles, the Supreme Court held that, to establish that an at-large, multi-member districting plan violates § 2 by “submerging” minority voters in a white majority, plaintiffs must prove three “threshold conditions”: (1) that the minority group “is sufficiently large and geographically compact to constitute a majority in a single-member district”; (2) that the minority group is “politically cohesive”; and (3) that “the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority’s preferred candidate.”16 Gingles, 478 U.S. at 46, 50-51, 106 S.Ct. at 2764, 2766-67; see also Shaw v. Reno, — U.S. -, -, 113 S.Ct. 2816, 2831, 125 L.Ed.2d 511 (1993) (reiterating the Gingles threshold conditions); Voinovich, — U.S. at-, 113 S.Ct. at 1157 (same); Growe v. Emison, — U.S.-,-, 113 S.Ct. 1075, 1084, 122 L.Ed.2d 388 (1993) (same).
In 1993 the Supreme Court twice held unanimously that the Gingles threshold conditions apply to § 2 vote dilution challenges to single-member, as well as multi-member, redistricting schemes. See Voino-vich, — U.S. at-, 113 S.Ct. at 1155-58 (analyzing a § 2 “packing” claim); Growe, — U.S. at-, 113 S.Ct. at 1084-85 (analyzing a § 2 “cracking” claim). Justice Sca-lia, speaking for a unanimous Court in Growe, explained the rationale for applying the three-part Gingles threshold inquiry to either multimember or single-member district systems:
*1045The “geographically compact majority” and “minority political cohesion” showings are needed to establish that the minority has the potential to elect a representative of its own choice in some single-member district.... And the “minority political cohesion” and “majority bloc voting” showings are needed to establish that the challenged districting thivarts a distinctive minority vote by submerging it in a larger white voting population.... Unless these points are established, there neither has been a wrong nor can be a remedy.
— U.S. at ——, 113 S.Ct. at 1084 (citations and footnote omitted; emphasis added).
The burden of proving the Gingles preconditions rests “squarely on the plaintiffs shoulders.” Voinovich, — U.S. at-, 113 S.Ct. at 1156; see also Gingles, 478 U.S. at 46, 106 S.Ct. at 2764 (“[T]he results test does not assume the existence of racial bloc voting; plaintiffs must prove it.”). The Fourth Circuit has clearly stated that plaintiffs cannot “ ‘pass the summary judgment threshold’ ” unless they establish all three Gingles preconditions. See McGhee v. Granville County, 860 F.2d 110, 117 (4th Cir.1988) (quoting McNeil v. Springfield Park Dist., 851 F.2d 937, 942 (7th Cir.1988), cert. denied, 490 U.S. 1031, 109 S.Ct. 1769, 104 L.Ed.2d 204 (1989)).
3. The Affirmative Defense of Proportional Representation. Defendants can generally defeat a § 2 claim by showing that the protected minority group has achieved or will achieve persistent proportional representation under the challenged districting plan. Proportionality defeats a § 2 vote dilution claim because a parity of votes and representation is compelling evidence that opportunities for the minority class to participate in the electoral process equal those of other members of the electorate.17
In Gingles, the State of North Carolina highlighted evidence that voters in House District 23, which was 36.3% black, had elected one black to their three-person House delegation in each of the previous six elections, spanning a twelve-year period. See Gingles, 478 U.S. at 74 & n. 35, 106 S.Ct. at 2778 & n. 35. Six Justices agreed that this “persistent proportional representation” based on the actual results of past elections was presumptively inconsistent with a § 2 violation. Id. at 77, 106 S.Ct. at 2780 (Brennan, J., joined by White, J.); id. at 102, 106 S.Ct. at 2793 (O’Connor, J., joined by Burger, C.J., and Powell and .Rehnquist, JJ., concurring on this point). Similarly, a State can defeat a § 2 vote dilution challenge by showing that the plaintiff class is likely to achieve proportional representation under a redistricting plan that has not yet been put into effect.
4. The “Totality of Circumstances” Inquiry. If the plaintiffs establish the three' Gingles threshold conditions and survive the affirmative defense of proportional representation, the District Court may consider other relevant factors in determining whether, under the “totality of circumstances,” minority group members have been denied an equal opportunity “to participate in the political process and to elect representa tives of their choice.” 42 U.S.C. § 1973(b); see Gingles; 478 U.S. at 80, 106 S.Ct. at 2781-82. The Senate Report accompanying the 1982 amendments to the Voting .Rights Act enumerated several factors that may support a § 2 vote dilution claim, such as a history of official discriminatory voting practices or of election campaigns characterized *1046by racial appeals. See S.Rep. No. 417, 97th Cong., 2d Sess., pt. 1, at 28-29 (1982) [hereinafter S.Rep.], reprinted in 1982 U.S.Code Cong. & Admin.News 177, 206-07, and quoted in Gingles, 478 U.S. at 36-37, 106 S.Ct. at 2758-59.18 In most cases, however, the evidence that plaintiffs introduce to establish the Gingles preconditions (and to defeat any showing of proportional representation) suffices to prove liability.19 As Justice O’Con-nor predicted in her concurrence in Gingles, “the basic contours of a vote dilution claim require no reference to most of the [factors that] were highlighted in the Senate Report. ... [Electoral success has now emerged, under the Court’s standard, as the linchpin of vote dilution claims_” Gin-gles, 478 U.S. at 92-93, 106 S.Ct. at 2788 (O’Connor, J., concurring in the judgment); see also Jenkins v. Red Clay Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1116 n. 6, 1135 (3d Cir.1993).
B. The Statewide Claim
1. Factual Overview. Maryland has forty-seven legislative districts. One Senator and three representatives to the House of Delegates are elected from each district. Under Maryland law a legislative district may elect all three Delegates at-large throughout the district; or the district may be subdivided into three single-member delegate districts, or into one single-member district and one two-member, at-large district. See Md. Const. art. III, §§ 2-3; Md.State Gov’t Code Ann. § 2-201(c). Under the 1992 plan challenged here nine of the legislative districts are majority African-American. A tenth is a district in which an aggregate of African-Americans and Hispanics constitutes a majority. There are twenty-seven Delegates to be elected from the nine majority-black districts, and a single-member, majority-black delegate district has been created in the tenth majority-minority legislative district. Historically, African-American Delegates also have been elected from districts with black voting-age populations of less than 50%. For example, in 1990 six of twenty-five African-American Delegates were elected from such districts; four of those Delegates were elected from solidly white districts.
African-Americans constitute 23.5% of Maryland’s voting-age population and 24.9% of the State’s total population. Approximately 80% of the total black population live in the Baltimore metropolitan area, in Prince George’s County, and on the Eastern Shore. Anne Arundel, Charles, Howard, and Montgomery Counties also have significant African-American populations.
*1047The 1990 census evidenced that the overall population of Baltimore City had declined and that, as a result, it would necessarily lose representation in the General Assembly. Under the districting plan previously in effect, there were nine legislative districts in the City, entitling it to nine Senators and twenty-seven Delegates. After the 1990 census, if all legislative districts throughout the State had been drawn so that they were absolutely equal in population, there would have been only 7.23 districts within the City’s boundaries. The Governor’s Redistricting Advisory Committee made a decision that the loss in the City’s legislative representation should be minimized to the extent legally permissible. The plan that the Committee recommended — and that the Governor ultimately adopted — establishes eight legislative districts that lie entirely or largely within the City, plus a ninth district along a corridor running into adjacent Baltimore County where a substantial number of African-Americans have moved. Six of the nine districts have a majority-black voting-age population.20
Approximately 50% of the total and 48% of the voting-age population in Prince George’s County is black. The plan recommended by the Committee and adopted by the Governor provides for three majority-black districts and three majority-non-Hispanic-white districts. It also provides for a seventh district that is 51% minority and 40% black, measured by voting-age population.
The districting plan submitted by plaintiffs would create one more majority-black senatorial district and two more majority-black delegate districts in Baltimore City than does the state plan.21 Similarly, in Prince George’s and Montgomery Counties, plaintiffs’ plan would create two additional senatorial districts and four additional delegate districts that are majority-black.22
2. Analysis. Plaintiffs do not contend that the State’s plan fails to provide African-Americans with a realistic opportunity to elect candidates of their choice in fair proportions in Baltimore City and Prince George’s County. They clearly could not do so. In Baltimore City the total African-American population is 59% and the voting-age African-American population is 56%. The five majority-black districts created by the State’s plan comprise 63% of the eight City districts.23 Moreover, a ninth majority-black district is created along á corridor between Baltimore City and Baltimore County.24 *1048Likewise, as stated above, the creation in Prince George’s County by the state plan of three majority-black districts, three majority-non-Hispanic-white districts, and a seventh district that is majority-minority and 40% black in voting-age population closely parallels the evenly split population in the County.25
Thus, plaintiffs do not make a regional attack upon the Baltimore City and Prince George’s County districts. Instead, they argue that the state plan fails because it does not configure districts in Baltimore City and Prince George’s County and adjacent areas in such a way as to make the number of majority-black districts proportional to the total African-American population in the State.
Plaintiffs’ claim proceeds from an erroneous premise. They assert that a mere showing of a general pattern of racially polarized voting is itself sufficient to require that district lines be drawn to maximize the number of majority-black districts, at least up to a number constituting the same proportion that African-Americans constitute of the total state population.26 This proposition is without support in the Voting Rights Act or the cases construing it.
The last sentence of the Act (commonly known as the “Dole amendment” or “Dole proviso”) provides 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(b). As Justice O’Connor has succinctly stated in considering this clause, “[t]here is an inherent tension between what Congress wished to do and what it wished to avoid, because any theory of vote dilution must necessarily rely to some extent on a measure of minority voting strength that makes some reference to the proportion between the minority group and the electorate at large.” Gingles, 478 U.S. at 84, 106 S.Ct. at 2784 (O’Connor, J., concurring in the judgment). Certainly, it could not have been Senator Dole’s intent to prevent fair proportionality from being one of the factors taken into account in determining whether a violation of the Act has occurred (or whether a proposed remedy to a proven violation is appropriate). Gingles itself holds that “persistent proportional representation” of black voters by black legislators is a defense to a Voting Rights Act claim. Id. at 77, 106 S.Ct. at 2780 (opinion of Brennan, J.), at 102, 106 S.Ct. at 2793 (O’Connor, J., concurring on this point). Similarly, an effort, such as the very one made by the State of Maryland in this case, to assure that African-Americans have full equal opportunity to participate in the political process by creating the maximum number of districts (consistent with the “one person, one vote” principle) in a metropolitan area or other region where many African-Americans reside is entitled to deference by the courts.
No court, however, has simply ignored the Dole amendment and held, as *1049plaintiffs would have us do, that the Voting Rights Act requires that redistrieting be implemented according to a numerical formula intended to achieve proportional representation merely because, as a general matter, many voters tend to vote for candidates of their own race. Indeed, the Fourth Circuit has expressly held that the Dole proviso “prevents a court from using proportional representation as the ultimate standard for assessing the legal adequacy of a remedial legislative redistricting plan.” McGhee, 860 F.2d at 118; see also Hines v. Mayor & Town Council of Ahoskie, 998 F.2d. 1266, 1274 (4th Cir.1993); Smith v. Brunswick County, 984 F.2d 1393, 1400 (4th Cir.1993); Nash v. Blunt, 797 F.Supp. 1488, 1499 (W.D.Mo.1992) (three-judge, court), aff'd mem., — U.S. -, 113 S.Ct. 1809, 123 L.Ed.2d 441 (1993).27 Therefore, as to plaintiffs’ statewide Voting Rights Act claim, we conclude that the defendants are entitled to judgment as a matter of law.
C. The Eastern Shore Claim
1. The State Plan. The NAACP plaintiffs also have alleged that the State’s dis-tricting plan for the House of Delegates, as it applies to the Eastern Shore, violates § 2 of the Voting Rights Act by impairing the opportunity of black voters “to participate in the political process and to elect representatives of their choice.” 42 U.S.C. § 1973(b).28 Under the state plan, the lower Eastern Shore is covered by Legislative Districts 37 and 38. District 37 contains all of Dorches-ter County, most of Caroline and Talbot Counties, and the western portion of Wicomi-co County. District 38'contains the eastern portion of Wicomico County (including the county seat of Salisbury), as well as the entirety of Somerset and Worcester Counties. More than 50,000 African-Americans reside in the two districts combined, comprising approximatély 25.4% and 22.0% of the total populations of Districts 37 and 38, respectively. Half of the black population in the two-district region is in Wicomico and Dorchester Counties. Although more blacks live in District 37 than in District 38, the largest concentration of blacks anywhere on the Eastern Shore can be found in the western part of Salisbury, just barely inside the border of District 38. Under the state plan, all Delegates from Districts 37 and 38 would be elected at-large. Maryland’s “Resident Delegate” statute requires that the three Delegates from District 37 reside in three different counties, and the same is true for District 38. See Md.State Gov’t Code Ann. § 2-201(d).29
*1050Districts 37 and 38 in the challenged plan are quite similar to their predecessors in the plans that have applied to every state legislative election for the past twenty years. In those two decades, and indeed throughout the two centuries since the State’s founding, these districts have not elected a single black to the Maryland Senate or House of Delegates. See 42 U.S.C. § 1973(b) (courts may consider “[t]he extent to which members of a protected class have been elected to office”). Furthermore, it is undisputed that no black candidate preferred by black voters on the Eastern Shore is likely to win a seat in the House of Delegates under the State’s plan.
Plaintiffs contend that the State’s district-ing plan impairs their ability to elect Delegates of their choice by first “cracking” or splitting a politically cohesive black community between Districts 37 and 38, and then “submerging” each piece of that community in a majority white, at-large district.30 Plaintiffs further contend that a sufficiently large and compact single-member delegate district with a majority-black voting-age population could be carved out of Districts 37 and 38. Specifically, the NAACP’s proposed delegate “District 54-9”31 would unite black areas of Dorchester and western Wicomico Counties, which are within the State’s District 37, with heavily black precincts in the western and downtown parts of Salisbury, which the State has placed in District 38.32 Because more than 95% of the land area and 82% of the population of proposed District 54-9 lie within the State’s District 37, the NAACP’s claim is closely akin to the plaintiffs’ claim in Crin-gles. In both cases, a black community claimed to have been “submerged” in a multi-member district that had a white majority. Cf. Gingles, 478 U.S. at 46-51, 106 S.Ct. at 2764-67. But because the plaintiffs here also complain that the District 37-District 38 line bisecting Wicomico County severs the black neighborhoods of Salisbury from the black areas in western Wicomico and Dorchester Counties, this case presents a “cracking” claim, too. Cf. Growe, — U.S. at -, 113 S.Ct. at 1084-85.33
By uniting the black community of Salisbury, western Wicomico County, and Dor-chester County, and lifting it out of an at-large, three-member district, the NAACP hopes to enhance the opportunity of black voters on the lower Eastern Shore to participate in the political process and to elect a Delegate of their choice. If plaintiffs’ hopes are realized, the black community that comprises roughly one-fourth of the population of the lower Shore might — for the first time in Maryland’s history — send a black Delegate to Annapolis as part of the lower Shore’s six-member House delegation. Thus, plaintiffs’ requested relief falls far short of proportional representation and does not implicate the questions of proportionality that required us to deny plaintiffs’ claims regarding Prince George’s County, Baltimore City, and Baltimore County. See supra Part V.B.
2. The NAACP’s Proposed “District 5U-9. ” According to the 1990 census, proposed District 54-9 contains 32,650 people, 56.4% of whom are black. Of its 23,660 residents of voting age, 53.6% are black.34 In Dorchester *1051County, the proposed district contains all of the municipalities of Vienna and Hurlock, and parts of Cambridge; in Wicomico County, it contains all of Hebron and parts of Salisbury and Fruitland. From roughly northwest to southeast (its greatest span), District 54-9 measures about thirty-two miles in length. In most places, District 54-9 is at least five miles wide. Except for a few spots where it follows municipal boundaries, the district’s narrowest corridor is about two miles across.
S. “Sufficiently Large and Geographically Compact.” The first Gingles precondition requires us to determine whether the black population within proposed District 54-9 is sufficiently large and geographically compact to constitute a majority in that district. See Gingles, 478 U.S. at 50-51, 106 S.Ct. at 2766-67. Because African-Americans comprise well over 50% of both the total population and the voting-age population, one would expect there to be no question as to whether the black population was “sufficiently large.” However, defendants argue that the “sufficiently large ... to constitute a majority” language in Gingles ’ first prong is not to be read so literally. “Majority,” they suggest, refers to a “viable” or “effective” voting majority that has the potential to elect a representative of its choice — a potential that they claim is lacking here.35
Defendants’ argument is wrong as a matter of law. Their briefs rely on Voinovich, — U.S. at -, 113 S.Ct. at 1155, for the proposition that plaintiffs must create a district in which the minority group has “a good chance” of electing its candidates of choice. In fact, the defendants have excised a sliver of dictum from Voinovich and turned it on its head. Justice O’Connor, speaking for a unanimous Court, stated: “A politically cohesive minority group that is large enough to constitute the majority in a single-member district has a good chance of electing its candidate of choice, if the group is placed in a district where it constitutes a majority.” Id. (dictum). She did not define a “majority” as the smallest possible set of people who have a good chance of electing their candidate of choice. Rather, she merely stated a legal presumption: any politically cohesive group that does constitute a numerical majority in a district can be presumed to have a good chance of electing its candidates of choice.
Under the law of the Fourth Circuit, that presumption is irrebuttable: “[W]hen a racial (or language) minority becomes the voting majority in a single-member district, it is empowered with the vote in that district.” Smith v. Brunswick County, 984 F.2d at 1401; see also id. at 1402 (holding that “when black voters have equal access to the polls and in fact represent a majority of those eligible to vote in a majority of the election districts relevant to the governmental body at issue, the rights afforded by ... the Voting Rights Act are satisfied” (emphasis added)). Similarly, in reviewing the same districting plan that is under attack here, Chief Judge Murphy of the Maryland Court of Appeals observed that “[f]ederal courts have frequently ordered the creation of ‘minority’ districts in which racial minorities constitute barely more than a majority of the population.” Legislative Redistricting Cases, 331 Md. at 607-08, 629 A.2d at 663 (citing, inter alia, McGhee, 860 F.2d at 113 (51.8% black voting-age population sufficient); Jordan v. Winter, 604 F.Supp. 807, 814 (N.D.Miss.) (three-judge court) (52.8% black voting-age population sufficient), aff'd mem., 469 U.S. 1002, 105 S.Ct. 416, 83 L.Ed.2d 343 (1984); Burton v. Sheheen, 793 *1052F.Supp. 1329, 1354 (D.S.C.1992) (three-judge court) (“[A] district may be considered a black opportunity district if the percentage of the black voting age population is greater than 50 percent.”), vacated and remanded, — U.S. -, 113 S.Ct. 2954, 125 L.Ed.2d 656 (1993)).36 Accordingly, we find that the black voting-age population within District 54-9 is “sufficiently large” under Gingles.
Next we must consider whether the NAACP’s proposed District 54-9 is “sufficiently ... geographically compact.” Gingles, 478 U.S. at 50, 106 S.Ct. at 2766. We begin by noting that compactness is neither mentioned in the text of the Voting Rights Act nor required by the federal Constitution. See Shaw, — U.S. at -, 113 S.Ct. at 2827 (O’Connor, J., for a five-member majority) (citing Gaffney, 412 U.S. at 752 n. 18, 93 S.Ct. at 2331 n. 18); id. — U.S. at -, 113 S.Ct. at 2841 (White, J., joined by Blackmun and Stevens, JJ., dissenting) (quoting Gaffney, 412 U.S. at 752 n. 18, 93 S.Ct. at 2331 n. 18); id., — U.S. at -, 113 S.Ct. at 2849 (Souter, J., dissenting).
The majority opinion in Gingles, from which the compactness requirement flows, unfortunately provides little guidance. Justice Brennan’s opinion for the Court explained that the compactness requirement was designed to bar § 2 suits where no majority-minority district can be drawn because the plaintiff minority group members are “substantially integrated” or “spread evenly” throughout the challenged district. See 478 U.S. at 50 & n. 17, 106 S.Ct. at 2766 & n. 17.37 Thus as originally described by Justice Brennan, Gingles’ compactness requirement simply precluded a finding of liability under § 2 where no remedy was possible.38
In more recent cases, however, the Supreme Court has clearly indicated that the concept of compactness is not a hollow one. In Growe v. Emison, Justice Scalia referred to a state senate district that “stretch[ed] from south Minneapolis, around the downtown area, and then into the northern part of the city in order to link minority populations” as an “oddly shaped creation” of “dubious” geographic compactness. Growe, — U.S. at -, -, 113 S.Ct. at 1083, 1085 (dicta). And, although its holding was grounded in the Equal Protection Clause rather than the Voting Rights Act, the Court in Shaw v. Reno, certainly focused its attention on the non-compactness of North Carolina’s Twelfth Congressional District. See Shaw — U.S. at ---, 113 S.Ct. at 2822-31. But see id., at -, 113 S.Ct. at 2831 (expressly reserving the question whether the district at issue was “geographically compact” within the meaning of Gingles).
From these recent cases we discern two guiding principles. First, courts *1053should be reluctant to order the creation of Voting Rights districts of “bizarre” or “dramatically irregular” shape. Id., at - —, -, -, 113 S.Ct. at 2820, 2825, 2831. Second, although a State can — and at times must — place great weight on race when redistricting, it may not do so to the exclusion of all traditional, nonracial districting principles, leaving a district that rationally can be understood only as “an effort to classify and separate voters by race.” Id. at -, 113 S.Ct. at 2828; cf. Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265, 287-320, 98 S.Ct. 2733, 2746-64, 57 L.Ed.2d 750 (1978) (plurality opinion). Put differently, the case law suggests that we must pay attention to both geometric and substantive criteria when testing for compactness under Gingles.
Whenever possible, Voting Rights litigants should adduce evidence systematically comparing proposed redistricting plans to either the plan being challenged or the predecessor plan in effect during the last relevant election. Social scientists have developed a plethora of methodologies for objectively measuring the compactness of individual districts and of entire plans,39 but we have no occasion to evaluate those techniques here, as none were put into evidence.
We are not, however, completely lacking in objective measurements that can serve at least as a proxy for the more sophisticated methods. The essence of defendants’ argument is that the NAACP’s proposed district takes two distant pockets of dense black population — one in Salisbury, the other in Cambridge — and strings them together with a narrow rural corridor. The argument, however, cannot withstand scrutiny. District 54-9 is only thirty-two miles long at its greatest span. Under the State’s plan, however, fourteen Delegates and nine Senators from around the State will be forced to serve constituents who are spread over a greater distance than District 54-9’s thirty-two miles.40 Cf. Jeffers v. Clinton, 730 F.Supp. 196, 207 (E.D.Ark.1989) (three-judge court) (“[Plaintiffs’] alternative districts are not materially stranger in shape than at least some of the districts contained in the present apportionment plan.”), aff'd mem., 498 U.S. 1019, 111 S.Ct. 662, 112 L.Ed.2d 656 (1991); Legislative Redistricting Cases, 331 Md. at 591-92, 629 A.2d at 654-55 (noting that a challenged district’s “shape, while unusual, is no more odd than the rest of the districts ... in the whole State”). The defendants also complain that the NAACP’s proposed districts use attenuated “corridors” — sometimes only two miles across — to link noncontiguous pockets of denser population. However, Voting Rights case law indicates that there is nothing extraordinary about this technique. See, e.g., Neal v. Coleburn, 689 F.Supp. 1426, 1435 (E.D.Va.1988) (holding that plaintiffs satisfied Gingles’ geographic compactness criterion even where “a number of fairly small pockets [of black population had to] be connected to create two election districts”). Indeed, the State apparently used a similar technique when drawing its own plan, as seven of its districts contain corridors less than half a mile wide. One passageway in the state plan’s District 18 is less than 200 yards across.
In short, if District 54-9 is not sufficiently compact, then the same can be said of many districts in the State’s new legislative redistricting plan. That plan, however, has already withstood a challenge brought specifi*1054cally on the ground of non-compactness. Only a few months ago, the Maryland Court of Appeals heard — and rejected — a claim that the State’s plan violated Maryland’s constitutional requirement that legislative districts “be compact in form.” Md. Const. art. III, § 4; see Legislative Redistricting Cases, 331 Md. at 580, 590-92, 629 A.2d at 648, 654-55. We see no reason to reach a different result.41 Therefore, we conclude that plaintiffs have met the burden of proving that proposed District 54-9’s shape and appearance are unobjectionable under Gingles.
Next, we turn to the question of whether District 54-9’s shape rationally can be understood only as “an effort to classify and separate voters by race.” Shaw, — U.S. at -, 113 S.Ct. at 2828. As an initial matter, we note that the NAACP drew many illustrative single-member districts on the Shore that had a significantly greater black population than District 54-9.42 The NAACP drew one district that contained 25% more voting-age blacks than District 54-9. Thus, District 54-9 could not have been merely the result of an effort to maximize the number of black voters and to minimize the number of white voters in the proposed district. Other considerations must have come into play.
Indeed, upon closer inspection, District 54-9 evidences considerable regard for traditional, nonracial redistricting criteria. See Shaw, at ---, 113 S.Ct. at 2822-32 (repeatedly invoking “traditional districting principles”). We begin by turning to those criteria set forth in the Maryland Constitution.43 First, District 54-9 is “of substantially equal population” with the plan’s other single-member districts, as it falls within four percent of the average delegate district population. Md. Const. art. III, § 4; Legislative Redistricting Cases, 331 Md. at 600-01, 629 A.2d at 659; see also supra Part III.
Second, the district “consistís] of adjoining territory.” Md. Const. art. III, § 4.44 And unlike the North Carolina district challenged in Shaw, discrete pieces of District 54-9 are never merely “point contiguous,” ie., touching only at a common corner. See Shaw, — U.S. at -, 113 S.Ct. at 2821.45
Third, District 54-9 gives “[d]ue regard ... to natural boundaries.” Md. Const. art. III, § 4. In interpreting this constitutional *1055provision, the Maryland Court of Appeals has stated that it is preferable to avoid water crossings where there is no bridge or ferry. See In re Legislative Districting, 299 Md. at 682, 475 A.2d at 440. District 54-9 straddles the Wicomico River through much of Wicom-ico County, but the river presents no travel difficulties, as the district encompasses both the Upper Ferry crossing to the west of Salisbury and the Route 50 crossing in Salisbury itself. District 54-9 also crosses the Nanticoke River, as it must, since that river constitutes the entire border between Dor-chester and Wicomico Counties, and the district was drafted to incorporate the Route 50 bridge that crosses the Nanticoke at Vienna.46
Fourth, the Maryland Constitution provides that each legislative district shall elect one Senator and three Delegates, and that the Delegates’ districts must be “nested” in, or coterminous with, the Senator’s. See Md. Const. art. III, § 3; see also Md. State Gov’t Code Ann. § 2-201(c). District 54-9 could be perfectly nested in the State’s Legislative District 37 if part or all of nine Wicomico County precincts were shifted from District 38 to District 37.47 In addition to complying with the nesting requirement, the NAACP’s proposal preserves the cores of both lower Shore senatorial districts and of five of the six lower Shore delegate seats.48
Fifth, District 54-9 gives “[d]ue regard ... [to] the boundaries of political subdivisions.” Md. Const. art. III, § 4. Under the State’s new plan, and under the plan that applied to the last three legislative elections, only three counties on the lower Shore' — -Dorchester, Somerset, and Worcester — are contained entirely within a single legislative (senatorial) district. Wicomico County is divided between Districts 37 and 38,49 and Caroline and Talbot Counties are divided between Districts 37 and 36. Nothing in the NAACP’s proposal would disturb that arrangement.50 District 54-9 also respects town limits. Of the six municipalities it touches, three are included entirely within the district and three are divided.51 Not surprisingly, the district divides the two largest towns that it touches- — Salisbury52 and Cambridge. In*1056deed, the only way to draw a single-member delegate district that includes all of Salisbury and Cambridge, without making the district unconstitutionally large in population, would be to connect the two cities with a razor-thin, thirty-mile-long rural corridor, so that the district would resemble an elongated dumbbell — a result that clearly would lack compactness. But that is not what the NAACP did here.53
Sixth and finally, District 54-9 “allows for effective representation.” Dillard v. Baldwin County Bd. of Educ., 686 F.Supp. 1459, 1466 (M.D.Ala.1988); see also Burton v. Sheheen, 793 F.Supp. at 1356. Samuel Q. Johnson, III, who is currently District 37’s Resident Delegate for Wicomico County and the Chairman of the House Delegation for Dor-chester County, testified at trial that in his current position he was perfectly capable of staying in touch with his constituents throughout western Wicomico County and the entirety of Dorchester County — an area considerably larger than District 54-9. Johnson, a three-term Delegate, stated that he could readily campaign in and represent District 54-9, and added, “It would save me a lot of mileage.” Thus, direct testimony from the State official most familiar with the territory contained in District 54-9 confirmed that the creation of a single-member district would facilitate, rather than hinder, effective political representation. See generally Pamela S. Karlan, Maps and Misreadings: The Role of Geographic Compactness in Racial Vote Dilution Litigation, 24 Harv.C.R.-C.L.L.Rev. 173 (1989) (arguing that a functional approach to Gingles’ geographic eom-pactness requirement is best suited to the inclusive spirit of § 2).
Therefore, we conclude that District 54-9 is not only compact in its shape and appearance, but moreover reflects the reasonable balancing of numerous legitimate redistricting principles: equal population, contiguity, due regard for natural and political boundaries, nesting, preservation of the cores of prior districts, effective representation — and equalizing the opportunities of all citizens, white and black, to elect representatives of their choice. Thus, plaintiffs have met the first prong of Gingles’ threshold inquiry.
4. Black Political Cohesion. The second Gingles precondition requires us to determine whether the black population is politically cohesive. See Gingles, 478 U.S. at 51, 106 S.Ct. at 2766-67. The Gingles Court stated that a “showing that a significant number of minority group members usually vote for the same candidates is one way of proving the political cohesiveness necessary to a vote dilution claim.” Id. at 56, 106 S.Ct. at 2769 (emphasis added). Courts and Voting Rights experts have interpreted “a significant number” to mean that minority support for minority candidates running against non-minority candidates reaches “landslide proportions” — typically defined as three-fifths of the total vote in a contest for a single position. See, e.g., Jenkins v. Red Clay, 4 F.3d at 1127; Allan J. Lichtman & J. Gerald Hebert, A General Theory of Vote Dilution, 6 La Raza L.J. 1 (1993).54
In Wicomico County, the plaintiffs easily showed black political cohesion. In three single-office contests, black candidates run*1057ning against white candidates garnered 71, 81, and 89 percent of the vote in heavily black precincts. On three other occasions, black candidates competed for seats in elections where voters could select more than one candidate. In a 1990 election for the Wicomico County Council, which involved three white candidates and one black candidate, voters could cast ballots for up to two candidates; yet black voters gave 84% of their votes to the black candidate — a sure indication that blacks were “single-shot” (or “bullet”) voting, ie., voting only for the black candidate while declining to cast their second ballot. Cf. Gingles, 478 U.S. at 87, 106 S.Ct. at 2785 (O’Connor, J., concurring in the judgment) (single-shot, or bullet, voting “occurs when voters refrain from easting all their votes to avoid the risk that by voting for their lower ranked choices they may give those candidates enough votes to defeat their higher ranked choices”); id. at 38 n. 5, 106 S.Ct. at 2760 n. 5 (majority opinion) (providing an example of bullet voting). Similarly, an African-American candidate running against four white opponents captured 68% of the black vote in a race where voters could cast up to three votes, and another black candidate (who ran in a field of 13) garnered 41% of the black Vote in a race where voters were permitted to select up to five candidates. Every one of these elections indicates not only overwhelming bloc voting on the part of the black community, but also a consistent effort to cast ballots exclusively for the black candidates. Cf. Cruz Gomez v. City of Watsonville, 863 F.2d 1407, 1415 (9th Cir.1988) (citing evidence of single-shot voting to reinforce a finding of minority political cohesion), cert. denied, 489 U.S. 1080, 109 S.Ct. 1534, 103 L.Ed.2d 839 (1989). In order to maximize their likelihood of electing black candidates, black voters were willing effectively to abandon their right to influence which white candidates would win.
The direct statistical evidence of black political cohesion in Wicomico County is corroborated by the results of actual elections in majority-black districts. After the U.S. Department of Justice filed a Voting Rights suit, Wicomico County created a single-member district for its County Council. In 1990, in the very first election held in Wicomico County Council District One — whose voting-age population is only 59% black — an African-American, Rudolph C. Cane, triumphed. In the town of Salisbury, legal action led to the creation of a majority-black district, but its population was only 58% black. Nevertheless, it promptly elected Gertrude Shockley, an African-American, in 1990. Thus, in the only two elections ever held in majority-black districts within Wicomico County, blacks were sufficiently cohesive to elect a black candidate, even though both districts contained a substantial white minority.
Dorchester County provides even stronger evidence of black political cohesion. In the two races pitting one black candidate against one white candidate, the black vote was virtually unanimous behind the former. And in the three black-versus-white races where each voter could cast up to three ballots, black candidates garnered an estimated 84, 93, and 100 percent of the black vote.55 The evidence of black bloc voting — and of black single-shot voting — is incontrovertible.
Again, the results of elections conducted in Dorchester' County’s majority-black districts corroborate plaintiffs’ statistical proof. Through most of the 19th century, Cambridge elected its city council at-large and no black was ever elected. In 1896 the city of Cambridge replaced its at-large electoral system with five single-member districts. Until recently, one of these five districts, known as “Ward 2,” had a population that was more than 90% black; it always elected a black couneilmember.56 In the late 1980s, as a result of a lawsuit filed by the U.S. Department of Justice, Ward 2 was “unpacked,” *1058decreasing its black population to about 70%, and a second majority-black district (with an approximately 60% black population) was created. Ward 2 continued to elect the same black councilmember, and in 1992 he was joined by another black city councilmember, from the 60% black Ward 3. In Hurlock, following a 1991 Voting Rights suit, the town created a majority-black district, which in turn elected Hurlock’s first black city coun-cilmember. And at the countywide level, Dorchester County elected a black, Lemuel Chester, as Commissioner from its newly-created, 79% black, single-member district in 1986 and again in 1990.57 All told, over the past century, fifty-six separate elections have been held in majority-black districts in Dor-chester County, and fifty-three times a black candidate has prevailed. That is minority political cohesion.
Furthermore, the black electorate in Dor-chester County is cohesive with the black electorate in Wicomico County. Two of the races that plaintiffs’ expert subjected to statistical analysis covered the entire area contained in the NAACP’s proposed District 54-9.58 In those two races, the black candidates received more than 85% of the total votes cast by black voters, while the white candidates got less than 15%. Compare Gingles, 478 U.S. at 81-82, 106 S.Ct. at 2783 (finding minority political cohesion even where the black vote in black-versus-white elections was divided, on average, 69% to 31%).59 Thus, there is no doubt that blacks throughout District 54-9 are politically cohesive and have a “distinctive minority vote.” Growe, — U.S. at -, 113 S.Ct. at 1084.
Defendants cite Growe60 for the proposition that this Court must examine black political cohesion not only within the proposed District 54-9, but within the entirety of District 37 (ie., all of Dorchester, plus the parts of Wicomico, Caroline, and Talbot Counties that fall within District 37). If plaintiffs’ expert had statistically analyzed Caroline and Talbot Counties’ election results, defendants surmise, he might have found black political cohesion behind black candidates there to be so low that the district-wide average of the four counties would indicate an absence of sufficient black cohesion.
Defendants’ argument is flawed for three reasons. First, Growe was concerned solely with a “cracking” claim in the context of a single-member districting scheme. Here, in contrast, plaintiffs assert that the plan “submerges” a politically cohesive black community in two at-large delegate districts where the white majority consistently votes as a bloc to defeat minority-preferred candidates. Thus, plaintiffs’ claim is essentially a “submergence” claim that is best analyzed by direct application of the Gingles test for mul-timember districts.
Second, nothing in Gingles or its progeny requires that there be cohesion between two geographically distinct groups that could not be placed together in a compact single-member district. See Gingles, 478 U.S. at 50-51, 106 S.Ct. at 2766-67. Indeed, to impose such a requirement would empower a State to defeat the very purpose of the Voting Rights Act by placing in the same multimember district two relatively distant minority populations that have no tradition of voting with one another and that *1059might have different interests, e.g., urban and rural blacks.
Third, assuming arguendo that defendants’ point has legal merit, it fails on the facts. Caroline and Talbot Counties have much smaller black voting-age populations than Dorchester or Wicomico County. As we noted above, the statistical evidence of black political cohesion in Dorchester and Wicomico Counties is powerful. For us to find that District 37, taken as a whole, lacks black political cohesion, we would have to assume that statistical evidence would have shown that blacks in Caroline and Talbot Counties usually cast less than 15% of their votes for black candidates and more than 85% of their votes for white candidates in black-versus-white elections—precisely the converse of what we found elsewhere on the Shore. Such an assumption strains credulity. Cf. Gingles, 478 U.S. at 45, 75-76, 79, 106 S.Ct. at 2763-64, 2779, 2781 (requiring trial courts to conduct an “independent consideration of the record” and a “searching practical evaluation of the past and present reality” (citation and internal quotation marks omitted)). Furthermore, such a finding would be hai'd to reconcile with the fact that the only majority-black district in Caroline or Talbot County (the Easton Town Council’s Ward 4, which is 65% black) has been represented for four of the last five years by an African-American.
5. Legally Significant White Bloc Voting. The third Gingles precondition for § 2 minority vote dilution claims requires us to determine whether the white majority votes sufficiently as a bloc to enable it usual-, ly to defeat the minority’s preferred candidate. See Gingles, 478 U.S. at 51, 106 S.Ct. at 2766-67. “The third prong of the test, the usual loss experienced by minority candidates, is a requirement that the nature of the dilution is real, not hypothetical.” Grofman & Handley, supra note 36, at 354 (emphasis added) (citing Gingles, 478 U.S. at 48, 106 S.Ct. at 2765).
Blacks in Wicomico, Dorchester, Caroline, and Talbot Counties rarely run for public office in majority-white constituencies, and when they do, they usually lose. At the county level, no black has ever been elected to any of the countywide single-member offices (ie., State’s Attorney, Clerk of Court, Register of Wills, or Sheriff). With only one exception, the four counties have never elected a black eouncilmember or commissioner at-large: Wicomico elected a black Republican County Councilmember, Emerson Holloway, in 1978, but he served just one term. The only other black public officeholders who have been elected countywide are Judge Norma Lee Barkley, who ran unopposed in the general election for one of three judge-ships on the Wicomico County Orphans’ Court in 1982 and was reelected in 1986 (on the strength of black single-shot voting) and again in 1990, and Judge James E. Thomas, who was elected to the Talbot County Orphans’ Court in 1986 but was defeated in 1990.
Since 1986, Dorchester and Wicomico Counties have subdistricted their county commission and county council, respectively, creating majority-white, as well as majority-black, single-member districts. Today, a black candidate has yet to win in any of those majority-white districts. Similarly, at the municipal level, blacks have met with little success in their occasional attempts to win public office either at-large or from majority-white districts.61
In Dorchester County, black candidates running against white candidates in the last seven years have received between zero and twelve percent of the white vote. Not surprisingly, no black has ever won public office countywide, at-large, or in a majority-white district.62 In Wicomico County, on average, less than 11% of white voters cast their *1060ballots for black candidates.63 Compare Gingles, 478 U.S. at 80-82, 106 S.Ct. at 2781-83 (finding legally significant white bloc voting in North Carolina even where, on average, more than one-third of the white electorate voted for black candidates).
Taken together, plaintiffs have shown both black political cohesion and legally significant white bloc voting — the two halves of racially polarized voting under Gingles. See id. at 52-74, 106 S.Ct. at 2767-79. At trial, those showings were verified by the testimony of a practical politician intimately familiar with District 37. The defendants called to the stand Delegate Samuel Q. Johnson, III, who has run successfully at-large in that district three times and who currently serves as both the Resident Delegate for western Wicomico County and the Chairman of the House Delegation for Dorchester County. Plaintiffs’ counsel asked Johnson if he would be willing to run in a majority-black, single-member district. Johnson initially said he would, but then asked how large the black majority would be. In a 90% black district, he explained, he might not even attempt a campaign, as “the NAACP will most likely make it a black-white race.” Counsel asked if he would be comfortable running in proposed District 54-9, which has a 53.6% black voting-age population, to which Johnson replied, “How about 50-50 to make it fair?” Delegate Johnson’s keen awareness of how his electoral fortunes were tied to the racial makeup of the district only served to reinforce the hard statistical evidence of racially polarized voting throughout District 37.
In sum, the Gingles threshold inquiry clearly shows that a bloc-voting white majority on the Eastern Shore consistently defeats black candidates supported by a politically cohesive and geographically compact black community. Plaintiffs’ claim of vote dilution is anything but hypothetical.
6. Totality of Circumstances. Although the evidence of racially polarized voting is powerful enough to stand on its own, the Voting Rights Act commands us to examine the totality of circumstances before determining whether the State delegate dis-tricting plan for the Eastern Shore denies blacks an equal opportunity to participate in the political process and to elect representatives of their choice. See 42 U.S.C. § 1973(b). Thus, we turn to the relevant factors that were first articulated in White v. Regester, 412 U.S. at 766-70, 93 S.Ct. at 2339-41, and Zimmer v. McKeithen, 485 F.2d 1297, 1305-08 (5th Cir.1973) (en banc), aff'd sub nom. East Carroll Parish Sch. Bd. v. Marshall, 424 U.S. 636, 96 S.Ct. 1083, 47 L.Ed.2d 296 (1976) (per curiam), and were later endorsed by the Senate in the Judiciary Committee’s report accompanying the bill that amended § 2. See S.Rep., supra, at 28-29.
a. Lingering effects of discrimination. One relevant factor identified in the Senate report is the extent to which members of a minority group “bear the effects of discrimination in such areas as education, employment, and health, which hinder their ability to participate effectively in the political process.” S.Rep., supra, at 29, 1982 U.S.Code Cong. & Admin.News 206. In Dorchester and Wicomico Counties, blacks are almost twice as likely as whites to have never attended high school, while whites are nearly three times as likely as blacks to have a college degree. When the white unemployment rate was below four percent, the black rate hovered near ten percent. Although blacks account for more than 25% of the two-county population, they earn less than 15% of its aggregate income. Whites are three times more likely to earn $75,000 or more per year, while blacks are three-and-a-half times more likely to fall below the poverty line. Indeed, more than one out of four black families in Dorchester and Wicomico Counties lives in poverty, and many others are in constant risk of joining their ranks. Although Dorchester and Wicomico are among the most rural counties in the State of Maryland, 27% of black households there have no vehicle. One witness testified at trial that the literal inability to get to the polls on Election Day significantly hampered *1061black turnout.64 See S.Rep., supra, at 29 n. 114 (stating that the Supreme Court has “recognized that disproportionate educational, employment, income level, and living conditions arising from past discrimination tend to depress minority political participation, ... [and that] plaintiffs need not prove any further causal nexus between their disparate socio-economic status and the depressed level of political participation”).
b. History of official discrimination in voting. Although on a statewide basis Maryland’s voting rights record is in many respects an admirable one, we cannot turn a blind eye to the Eastern Shore’s “history of official discrimination” that impaired blacks’ rights to register and to vote. S.Rep., supra, at 28. In 1904 Maryland’s General Assembly enacted the “Poe amendment” to the state constitution, which would have effectively disenfranchised most black voters. In reviewing the State’s history, the defendants make much of the fact that, in a 1905 ratification referendum, Maryland’s voters soundly rejected the Poe amendment. The defendants fail to note that the General Assembly enacted the amendment the previous year at the insistence of legislators from the Eastern Shore who agreed to back an oyster-seeding measure in exchange for Western Shore legislators’ support for black disenfranchisement. Furthermore, when the Poe amendment was subjected to a statewide referendum, only three counties voted resoundingly in favor of the disenfranchising of Maryland’s blacks—and all three were on the lower Shore. See Robert J. Brugger, Maryland: A Middle Temperament 420-24 (1988); Margaret Law Callcott, The Negro in Maryland Politics, 1870-1912 at 114-15, 125 (1969).
Maryland’s discriminatory voting practices are not only found in the history books. Until 1988, Maryland law condoned “dual registration,” which required voters to register separately for municipal and non-municipal elections.65 African-American citizens who had been historically excluded from full participation in political life, and hence were unfamiliar with registration procedures, frequently were turned away at the polls because they had only registered for one type of election. The dual registration requirement confused voters, depressed turnout, and—according to a 1985 report by Attorney General Stephen H. Sachs—may have resulted in' the dilution of black voting strength on the Eastern Shore. See Office of the Attorney General, Ah-Large Election of County Commissioners, at 29, 110-11 (1985) [hereinafter Attorney General’s Report].
Second, at least until the mid-1980s, some all-white, but state-funded, volunteer fire departments on the Eastern Shore functioned as a kind of unofficial slating organization for white candidates. See Attorney General’s Report, supra, at 28 & n. 21; cf. United States v. Marengo County Comm’n, 731 F.2d 1546, 1569 (11th Cir.) (“[W]here there is an influential official or unofficial slating organization, the ability of minorities to participate in that slating organization and to receive its endorsement may be of paramount importance.”), appeal dismissed and cert. denied, 469 U.S. 976, 105 S.Ct. 375, 83 L.Ed.2d 311 (1984). Only in 1988, upon the Attorney General’s written recommendation, did the Governor amend the Code of Fair Practices to prevent racially discriminatory fire departments from receiving state funds. See Attorney General’s Report, supra, at 28 & n. 21; Code of Maryland Regulations (COMAR) 01.-01.1988.05. Even today, counties on the lower Shore continue to locate polling places in white-dominated volunteer fire companies, a hostile environment that may depress black electoral participation. See Trial Test, of Carl Snowden.
c. The Resident Delegate statute. The Senate Report also instructs us to consider “anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group.” S.Rep., supra, at 29, 1982 U.S.Code Cong. & Admin.News 206. Plaintiffs claim that Maryland’s “Resi*1062dent Delegate” statute effectively mandates a “designated-post” system that operates as an anti-single shot provision and thus dilutes black voting strength on the Shore. See City of Rome v. United States, 446 U.S. 156, 185 n. 21, 100 S.Ct. 1548, 1566 n. 21, 64 L.Ed.2d 119 (1980) (noting that single-shot voting could be frustrated, and minority voting strength diluted, by “requiring each representative] to live in a separate district but with voting still at large” (citation and internal quotation marks omitted)); see also Lockkart v. United States, 460 U.S. 125, 134-35, 103 S.Ct. 998, 1004, 74 L.Ed.2d 863 (1983) (recognizing that a designated-post system “may frustrate the use of ‘single-shot voting’” and may “have the effect of discriminating against minorities ... where racial bloc voting predominates”).66
Black voters often have resorted to “single-shot” (or “bullet”) voting in at-large elections to overcome the handicap of being submerged in a bloc-voting white majority. Essentially, when a minority group engages in single-shot voting, it concentrates its votes behind one candidate and hopes that the white majority will divide its votes among several competing candidates. The 1986 Democratic primary for Wicomico County Orphans’ Court provides a perfect example of successful single-shot voting. Incumbent Judge Barkley, an African-American, was competing with four white candidates for three judgeships. Each voter could cast up to three votes. White voters gave 85% of their votes to the four white candidates and 15% to Judge Barkley, placing her in a tie for last place among white voters. Black voters, however, gave Judge Barkley 68% of their votes. Clearly, they were single-shot voting, ie., voting once for Judge Barkley and withholding their other two votes; otherwise, even if every black voter had supported her, she could not have received more than a third of the total black vote.
Maryland’s Resident Delegate statute requires candidates to run for designated posts 0e.g., for the Wicomico County seat, or for the Dorchester County seat, etc.) on the ballot, thus breaking what would otherwise be a single contest for three Delegates into three contests, each for only one Delegate.67 Each voter can cast only one vote per contest; thus by withholding votes, she is not taking votes away from the rivals of her preferred candidate. Therefore, the “designated post” scheme established by Maryland’s Resident Delegate statute frustrates single-shot voting and impairs the opportunity of Eastern Shore blacks to elect their Delegate of choice. See Chandler Davidson, The Voting Rights Act: A Brief History, in Controversies in Minority Voting 7, 25 n. 63 (Bernard Grofman & Chandler Davidson eds., 1992).
Defendants, however, argue that the Resident Delegate statute militates in favor of the state plan and against a finding of minority vote dilution. They claim that the State has a substantial, longstanding, and racially-neutral interest in attempting to assure each Eastern Shore county at least one Resident Delegate. Cf. Houston Lawyers’ Ass’n v. Attorney General of Tex., 501 U.S. 419, -, 111 S.Ct. 2376, 2381, 115 L.Ed.2d 379 (1991) (explaining that a State’s interest in a particular electoral practice should properly be considered when weighing the totality of circumstances in a minority vote dilution case). The counties of the lower Shore rely upon the General Assembly to enact public local laws, typically sponsored by their resident legislators. See Legislative Redistricting Cases, 331 Md. at 621-23, 629 A.2d at 670 (Eldridge, J., dissenting). See generally Bin-*1063ny Miller, Who Shall Rule and Govern? Local Legislative Delegations, Racial Politics, and the Voting Rights Act, 102 Yale L.J. 105, 128-31 (1992).
Maryland’s interest in its Resident Delegate system is not nearly so compelling as defendants would have us believe. First, under the State’s 1992 redistricting plan, almost a third of Maryland’s twenty-three counties — Calvert, Caroline, ' Dorchester, Garrett, Kent, Queen Anne’s, and Talbot— are left with no “guarantee” of representation in the House of Delegates. If the State’s interest in Resident Delegates were truly critical, the State could have made each one of those counties a single-member delegate district — thus ensuring each a Resident Delegate. See Brown v. Thomson, 462 U.S. at 843-48, 103 S.Ct. at 2696-99 (holding that, under the Fourteenth Amendment’s Equal Protection Clause, a consistently-applied state policy ensuring that each county has at least one representative in the state legislature can justify an 89% maximum deviation from population equality); see also Legislar tive Redistricting Cases, 331 Md. at 600-01, 629 A.2d at 659 (holding that the Constitution of Maryland does not impose a stricter standard for population equality than that imposed by the Fourteenth Amendment).
Second, even without redrawing the dis-tricting plan to make delegate districts and counties coterminous, the State could have virtually guaranteed each of the nine Eastern Shore counties at least one Resident Delegate. Based on 1990 census data, the nine counties on the upper and lower Shore could support a total of 10.13 Delegates if all districts in the State were equal in population. District lines could be drawn so that each of the nine counties dominated at least one single-member delegate district. Indeed, one of the plans submitted to the Governor’s Redistricting Advisory Committee would have given each county on the Shore at least one Delegate. The Committee rejected that plan, and the General Assembly, by its silence, acceded. The State provides no explanation of why an interest that was shunted aside during the redistricting process in 1991 has suddenly become so compelling.
Even assuming arguendo that the State’s interest were compelling, the NAACP’s proposed District 54-9 would do little, if any, damage to that interest. Under the State’s plan, Wicomico, Somerset, and Worcester Counties are each guaranteed a Delegate from District 38; that would remain unchanged under the NAACP’s proposal. And, under the State’s plan, Dorchester County would have either zero or one Delegate; under the NAACP’s proposal, Dorchester could end up with zero, one, or two Delegates. Today, under the State’s existing plan, Dor-chester County has no Resident Delegate, as District 37’s three Delegates live in Wicomi-co, Caroline, and Talbot Counties. Thus, the State cannot credibly claim that enactment of the NAACP’s proposal would significantly undermine the State’s interest in assuring county representation in the House of Delegates.
We conclude that the State has some legitimate interest in maximizing the number of counties that are represented in the state legislature by at least one county resident. See State Admin. Bd. of Election Laws v. Calvert, 272 Md. 659, 673-74, 327 A.2d 290, 298 (1974), cert. denied, 419 U.S. 1110, 95 S.Ct. 784, 42 L.Ed.2d 807 (1975); Maryland Comm, for Fair Representation v. Tawes, 229 Md. 406, 411-14, 184 A.2d 715, 717-19 (1962), rev’d on other grounds, 377 U.S. 656, 84 S.Ct. 1429, 12 L.Ed.2d 595 (1964). That interest, however, is far too tenuous to outweigh the fact that the Resident Delegate statute, as applied to District 37, exacerbates the dilution of black voting strength caused by the at-large election of Delegates from the lower Shore. Cf. S.Rep., supra, at 29 n. 117 (stating that even a practice that was premised on a racially neutral policy and that has been consistently applied may deny minorities fair access to the political process and thus violate § 2).
d. The failure of black candidates to win public office. In amending the Voting Rights Act in 1982, Congress found that poor socioeconomic conditions, a history of official discriminatory voting practices, electoral devices such as anti-single shot provisions, and — most important — severe and persistent racially polarized voting — could combine to *1064impair black candidates’ ability to win public office. See S.Rep., supra, at 29, 1982 U.S.Code Cong. & Admin.News 207 (highlighting “the extent to which members of the minority group have been elected to public office”); see also 42 U.S.C. § 1973(b). The Eastern Shore fits precisely that pattern. Although across the State, black candidates in Maryland have met with considerable success in recent years, unequal political opportunity has continued to plague the Eastern Shore. In Maryland’s two hundred year history, there has never been a black State Senator or Delegate from any of the nine counties of the Eastern Shore. Indeed, in the 123 years since ratification of the Fifteenth Amendment, only two blacks have even run for the House of Delegates from the lower Shore. The State’s recently enacted redistricting plan promises more of the same.
Out of literally hundreds of countywide elections for public office that have taken place in Wicomico, Dorchester, Caroline, Talbot, Somerset, and Worcester Counties, only five have resulted in a black candidate’s victory. And within the confines of District 37, only three African-Americans have been elected at-large to a municipal public office.
By contrast, black candidates have been highly successful in single-member districts with black voting-age majorities, even when the districts contained sizeable white populations. Across the entire lower Shore, there are now a total of twelve majority-black or Voting Rights districts. Eleven have elected a black candidate to represent them. Thus, the Voting Rights Act has already significantly improved the fairness of political representation on the Shore. A cadre of new black elected officials is gaining experience in the ways of county and municipal government. But, absent some relief, the doors to state government will remain closed both to this new generation of African-American leaders and, more importantly, to the people whom they represent.
We find that a bloc-voting white majority consistently defeats candidates supported by the lower Shore’s politically cohesive, geographically compact black community. And we find that, under the totality of circumstances, blacks on the Eastern Shore have less opportunity than whites to participate in the political process and to elect Delegates of their choice. Therefore, we hold that the State’s districting plan, as it applies to the Eastern Shore, violates § 2 of the Voting Rights Act, and we order the State to remedy that violation. In doing so, we hope to unlock doors that for too long have been closed to the African-American community on the Eastern Shore.
Accordingly, we will direct the State of Maryland to prepare' and submit to the Court by February 28, 1994,68 a new redistricting plan for the Eastern Shore that (1) creates a single-member delegate district with a majority-black voting-age citizen population, and (2) complies with the Constitution of the United States, with the Voting Rights Act of 1965, 79 Stat. 437, as amended, 42 U.S.C. § 1973, and with the Constitution and laws of the State of Maryland. A separate order effecting the rulings made in this opinion is being entered herewith.
*1065APPENDIX
[[Image here]]
*1066APPENDIX
[[Image here]]
. Judge Smalkin dissents from the majority’s conclusion that plaintiffs have proven a Voting Rights Act violation on the Eastern Shore.
. GRAC’s members were: Benjamin L. Brown, a prominent Baltimore attorney, chairman; Thomas V. "Mike” Miller, Jr., President of the State Senate; R. Clayton Mitchell, Jr., Speaker of the House of Delegates; Norman M. Glasgow, Sr., a Montgomery County attorney; and Donna M. Felling, a former member of the House of Delegates from Baltimore County. Chairman Brown is an African-American.
. These included: equality of population between districts, minority representation, contiguity, compactness, and respect for natural boundaries and the boundaries of political subdivisions. See Legal Standards for Plan Development at II.A.-F.
. These included: respect for established precinct lines, communities of interest, and existing district lines, and subdistricting to protect county integrity and to assure each county at least one Delegate, to the extent possible. See id. at II.G.J.
. Approximately 300 persons testified at those hearings. The GRAC also considered more than forty written submissions for proposed plans that it received from the public.
. More than 100 persons testified at that hearing.
. Article III, § 5 of the Maryland Constitution provides that if the General Assembly fails to adopt a redistricting plan of its own by the 45th day of its session in the second year after the census, the Governor’s plan becomes law.
. Racial discrimination or a state policy of purposefully disfavoring counties whose names begin with the letter "C," for example.
. This Court agrees with the Court of Appeals of Maryland that the plaintiff must be required to prove that no legitimate policy concerns contributed to the deviation or the very purpose of the ten percent rule would be defeated. See Legislative Redistricting Cases, 331 Md. at 599 n. 20, 629 A.2d at 658 n. 20. The ten percent rule permits states to incorporate legitimate state policies into redistricting plans without the burden of justifying the minor population deviations that the incorporation of those policies might cause. "Disgruntled opponents of any redistricting scheme can invariably offer something suggesting ill will by district planners; thus, [a failure to require the plaintiff to show that the promotion of no legitimate state policy contributed to the deviation] would undermine Brown’s rule that maximum deviations under 10% will not 'ordinarily' require justification by the State.” Id. If the plaintiff were not required to prove that no legitimate policy contributed to the minor deviation, every redistricting scheme "would languish in protracted litigation,” the very result the ten percent rule was designed to avoid. Id.
. In fact, the "Legal Standards” document to which the plaintiff has referred expressly, states that the Committee's goal was that the “Senate districts shall be substantially equal in population, taking into account constitutional and his*1035torical recognition of political subdivisions.” Legal Standards for Plan Development at II.A.2.(a).
. This inference is actually supported by the “Legal Standards” document cited by the plaintiff. The very document that provides a population window also addresses the promotion of other state policies such as minority representation, contiguity, compactness, and respect for political subdivisions, existing district lines, and incumbency. See Legal Standards for Plan Development, at II.B-J.
. The Court wishes to emphasize that although it is assuming, for analytical purposes only, that this objective is unconstitutional, it is far from convinced that the Constitution prohibits a State from making the political determination that a region’s or political subdivision’s representation should be maximized, so long as no individual’s vote is substantially diluted. From Reynolds v. Sims through the present, the emphasis of one person, one vote jurisprudence has been on the dilution of the votes of individuals. See, e.g., Reynolds, 377 U.S. at 568, 84 S.Ct. at 1385 (“Simply stated, an individual's right'to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the State.” (emphasis added)); Hadley v. Junior College Dist., 397 U.S. 50, 54, 90 S.Ct. 791, 794, 25 L.Ed.2d 45 (1970) ("This Court has consistently held in a long series of cases, that in situations involving elections, the States are required to insure that each person's vote counts as much, insofar as it is practicable, as any other person’s vote.” (emphasis added; footnote omitted)). The concern in one person, one vote decisions with a preference of one region to another has been in the context of dis-tricting plans that favor underpopulated regions, such as rural counties, by substantially diluting the individual voting strength of residents in more populated regions. See, e.g., Reynolds, 377 U.S. at 550, 84 S.Ct. at 1375-76 (voters in rural Alabama counties had fifteen to twenty times more voting strength than voters in highly populated counties); see also Abate v. Mundt, 403 U.S. 182, 185-86, 91 S.Ct. 1904, 1906-07, 29 L.Ed.2d 399 (1971); Hadley, 397 U.S. at 52-53, 90 S.Ct. at 792-94. In this case, MFR has not identified evidence that suggests that the weight of a resident's vote in any other region of the State is substantially diluted due to the eight Baltimore City Senate districts. As the Court of Appeals of Maryland noted, even Montgomery County, the only county that might possibly argue discrimination, controls precisely the number of state Senators that its population indicates it should control. See Legislative Redistricting Cases, 331 Md. at 598 n. 18, 629 A.2d at 658 n. 18. Moreover, it seems that MFR would still complain even if senatorial district lines were gerrymandered in such a way that all districts had equal populations, but Baltimore retained control over 8 districts. For this reason, perhaps the test established in the Bandemer decision, 478 U.S. 109, 106 S.Ct. 2797, 92 L.Ed.2d 85 (1986), for partisan gerrymandering is more appropriate than the one person, one vote decisions which concentrate on individual vote dilution. Because MFR would, in all probability, be unable to satisfy the demanding test established by Bandemer for discriminatory effect, see infra, it is far from clear that the GRAC’s objective here is unconstitutional.
. For example, the state plan distributes Democratic votes in both delegate and senatorial districts "inefficiently” by creating several districts in which the Democratic percentage of voters exceeds seventy or, in some cases, even eighty percent of the total voters. The plan establishes no Republican districts in which Republican votes are "wasted" to this extent. See Lichtman Aff. at ¶¶ 16, 26. In addition, the state plan pits no incumbent Republican Senators against each other or against incumbent Democrats while it pairs four Democratic incumbents. Id. at ¶ 20. No incumbent Republican Senator has been placed in a district that jeopardizes his or her chance of reelection. Id. at V 21. Similarly, only three Republican Delegates are paired exclusively in a two-member district as compared to the pairing of twenty-one Democrats in six districts. Id. at ¶ 28. These undisputed facts arguably negate any inference of discriminatory intent.
. Maryland has one of the highest concentrations of Democrats of any State in the nation, with Democrats constituting more than two-thirds of the total (two-party) registration. See Lichtman Aff. at ¶ 33 n. 9.
. The Voting Rights Act claims raised here are distinct from those considered and rejected by the Maryland Court of Appeals in Legislative Redistricting Cases, 331 Md. at 602-09, 629 A.2d at 660-63. The state-court petitioners raised no Voting Rights claims regarding the Eastern Shore. See id.
. Under the third prong of Gingles’ threshold inquiry, district courts typically consider experts' statistical analyses of census data and of past election returns to determine whether "a white bloc vote normally will defeat the combined strength of minority support plus white 'crossover' vote for the minority's preferred candidates.” Collins v. City of Norfolk, 883 F.2d 1232, 1237 (4th Cir.1989) (citing Gingles, 478 U.S. at 56, 106 S.Ct. at 2769), cert. denied, 498 U.S. 938, 111 S.Ct. 340, 112 L.Ed.2d 305 (1990).
. If proportional representation did not defeat a vote dilution claim, § 2 could mandate bizarre results. For example, imagine a city with (1) a five-person city council elected from five single-member districts, (2) a 40% black population, concentrated in one part of town, and (3) a highly racially polarized electorate. If the council district lines were drawn to create five "packed" districts, three of which were almost all white, and two of which were almost all black, one would expect the city council to consist of three white members and two black members, a result proportional to the city population. Nevertheless, black plaintiffs who sued under § 2 could easily establish the three Gingles threshold conditions. But it would be perverse for a federal court to find that the "packed” plan violated § 2 and to demand the creation of a third black-dominated district. The proportional representation defense precludes this scenario. In other words, the proportional representation defense constrains federal judicial power: in the above example, a federal court cannot force the city to draw more than two majority-black districts because doing so would result in unequal electoral opportunities for blacks and whites.
. A district court might consider the following factors:
1. the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;
2. the extent to which voting in the elections of the state or political subdivision is racially polarized;
3. the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;
4. if there is a candidate slating process, whether the members of the minority group have been denied access to that process;
5. the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;
6. whether political campaigns have been characterized by overt or subtle racial appeals; 7.the extent to which members of the minority group have been elected to public office in the jurisdiction.
Additional factors that in some cases have had probative value as part of plaintiffs' evidence to establish a violation are:
whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group [and]
whether the policy underlying the state or political subdivision’s use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.
Id. The Gingles Court noted that "this list of typical factors is neither comprehensive nor exclusive." 478 U.S. at 45, 106 S.Ct. at 2763.
. The Gingles Court explained that the so-called "Senate Report factors" — other than racially polarized voting and the failure of minority candidates to win public office — are supportive of, but not essential to, a § 2 vote dilution claim. See 478 U.S. at 48-49 n. 15, 106 S.Ct. at 2765-66 n. 15.
. The Committee’s plan (for Baltimore City) is substantially the same as a plan proposed by the Black Legislative Caucus. The only significant difference is that the Committee’s plan provides for at-large voting within one of the legislative districts whereas the plan proposed by the Caucus would have created one single-member delegate district with a 71.5% African-American population and one two-member delegate district with a bare African-American majority of 51.3%.
. Plaintiffs’ plan would increase the number of majority-black delegate districts by disregarding the requirement of the Maryland Constitution that delegate districts be "nested" within senatorial districts. See Md. Const. art. III, § 3. Plaintiffs are correct in arguing that state statutes or constitutional provisions that result in violations of the U.S. Constitution or the Voting Rights Act are preempted by federal law. See U.S. Const. art. VI, cl. 2; Md. Declaration of Rights, art. II; see, e.g., Hamm v. City of Rock Hill, 379 U.S. 306, 311-12, 85 S.Ct. 384, 389-90, 13 L.Ed.2d 300 (1964). On the other hand, it is self-evident that a state plan should — indeed must — comply with state statutory and constitutional mandates if compliance with those mandates does not violate federal law.
. The additional districts that plaintiffs' plan would create substantially decrease the percentage of African-Americans in most of the districts. Dr. Allan J. Lichtman, defendants' expert, is of the opinion that in at least some of those districts the decrease would actually result in the election of fewer candidates preferred by African-Americans. Plaintiffs' expert, Dr. Theodore S. Arring-ton, is of the contrary view. If the case turned upon resolution of their disagreement, disposition by way of summary judgment would, of course, be improper.
. Furthermore, plaintiffs cannot prove legally significant white bloc voting in Baltimore City: candidates preferred by Baltimore City blacks have won at least two-thirds of all recent black-versus-white elections in the City.
. Plaintiffs also could not contend that the State's plan fails to provide African-Americans with a realistic opportunity to elect candidates of their choice in fair proportions in the metropolitan region consisting of Baltimore City plus Baltimore County. The region is large enough to support fourteen equipopulous legislative districts (i.efourteen Senators and forty-two Delegates). Blacks comprise 36% of the total popula*1048tion and 34% of the voting-age population in the metropolitan region and therefore would be proportionately represented if they elected a total of five Senators and fifteen Delegates. Under the State's plan, however, the region will send six Senators and eighteen Delegates to Annapolis from majority-black districts.
. Furthermore, plaintiffs cannot prove legally significant white bloc voting in Prince George's County: candidates preferred by Prince George's County blacks have won more than two-thirds of all recent black-versus-white elections in the County.
. Plaintiffs now deny that they seek to maximize the number of African-American districts and assert that defendants' contention that they have done so is "incendiary.” However, in a memorandum submitted to the Governor's Redistricting Advisory Committee with the NAACP's proposed plan, Samuel L. Walters, Assistant General Counsel for the NAACP, stated that “[t]he current status of the law interpreting the Voting Rights Act makes it clear that where it is possible to draw majority black districts this must be done.” Further, on deposition plaintiffs' expert, Dr. Arrington, was unable to draw any distinction between the test that he applied in measuring the validity of the State’s plan and maximization of majority-black districts. After defendants filed their motion for summary judgment, Walters filed an affidavit stating that if maximization had been his goal, he could have drawn a plan containing more majority-black districts. But even by their own theory of the case, plaintiffs cannot justify more majority-black districts than they originally proposed because then the fraction of districts that are majority-black would substantially exceed the fraction of the State’s population that is African-American.
.Although we need not reach the issue, we also note that it appears that even assuming the validity of plaintiffs’ contention that the number of majority-black districts statewide should mirror the proportion of the State's African-American population, their claim would fail. There are nine majority-black senatorial districts and a tenth senatorial district that Dr. Arrington admits could at least “promote the development of mul-ti-racial coalitions.” Dr. Arrington further concedes that although eleven majority-black senatorial districts "would be a little better,” ten "would be close” to making the number of Senators elected by African-Americans proportional to the black voting-age population in the State. The percentage of majority-black delegate districts (20%, or 28 out of 141) is slightly less. However, the record establishes that several African-American Delegates have been elected from districts in which blacks are not a majority, including a few from solidly white districts. Whether these candidacies be viewed from the perspective of "group rights” or as evidence that the process of interracial coalition-building is working in at least some parts of the State, it would seem that in an analysis of statewide voting patterns, some consideration must be given to the election of black candidates from white districts.
Plaintiffs also assert statewide minority vote-dilution claims under the Fourteenth and Fifteenth Amendments to the U.S. Constitution. However, these claims require proof of essentially the same facts as do plaintiffs' Voting Rights Act claims, plus the additional element of discriminatory intent and purpose. See City of Mobile v. Bolden, 446 U.S. 55, 66-74, 100 S.Ct. 1490, 1499-1503, 64 L.Ed.2d 47 (1980) (plurality opinion).
. Plaintiffs have not challenged the plan as it affects the election of Senators from the Eastern Shore.
. Maryland’s Resident Delegate statute provides:
Residence of Delegates in districts containing more than 2 counties. — In any legislative district which contains more than 2 counties or parts of more than 2 counties, and where Delegates are to be elected at large by the voters of the entire district, a county, or part of a county, may not have more than 1 Delegate residing in that district.
*1050Md. State Gov't Code Ann. § 2-201(d). Under the State's redistricting plan, the Resident Delegate statute will apply only to the Eastern Shore.
. In addition, plaintiffs claim that the application of Maryland's Resident Delegate statute to Districts 37 and 38 further dilutes black voting strength by frustrating efforts to "single-shot” or "bullet" vote. See infra Parts V.C.4 and V.C.6.c.
. Consistent with the State's numbering system, District 54-9 would actually be denominated District 37A or 37B.
. In the course of the litigation, the NAACP proposed several other configurations, but we focus only on District 54-9, which, for illustrative purposes, presents plaintiffs’ claim most clearly.
. The line dividing Districts 37 and 38 in Wi-comico County follows no natural boundary.
. District 54-9's total population of 32,650 is below average for a single-member delegate district. Although the district itself complies with the Equal Protection Clause's "one person, one vote” requirement, see Voinovich, — U.S. at -, 113 S.Ct. at 1159, the remainder of District 37, which would be left if District 54-9 were removed, would be slightly too large for a two-member district (or for two single-member districts). There are two alternative cures for this defect. First, as plaintiffs suggested at trial, a few precincts of Caroline or Talbot County (both of which are already split between Districts 37 and 36) could be shifted from District 37 into District 36. Second, a few Wicomico County *1051precincts or partial precincts that are in District 37 but not in District 54-9 could be moved to District 38. District 54-9 then could be slightly enlarged, making the remainder of District 37 correspondingly smaller. For example, adding 2,000 persons to District 54-9 would (a) prevent the remainder of District 37 from becoming too large; (b) not threaten District 54-9’s essential attribute as a district with a black voting-age citizen majority (even assuming that all 2,000 persons added to the district were white); and (c) allow District 54-9's borders to be further smoothed out, thus bolstering its regularity and compactness.
. Defendants’ point is rather speculative. The record does not indicate how the residents of District 54-9 actually voted in any particular election. Nor would it be a simple task to estimate accurately how they voted in, say, the 1988 Democratic Presidential primary, because much of the proposed district’s border does not adhere to existing precinct lines.
. We need not address the question of whether the first prong of Gingles could be satisfied by a minority group that (a) comprises less than half of the voting-age population, but (b) nevertheless has a realistic potential to elect candidates of its choice (e.g., because the minority group has grown much more rapidly than the non-minority population since the last census was taken, because the minority voters are more cohesive than the non-minority voters, or because the minority voters would comprise a clear majority in the primary elections that determine the nominees for the district’s dominant party). We also need not address the question of whether "influence-district” claims are cognizable under § 2. See Voinovich, — U.S. at -, 113 S.Ct. at 1155; Growe, - U.S. at -, 113 S.Ct. at 1084 n. 5; Gingles, 478 U.S. at 46 n. 12, 106 S.Ct. at 2764 n. 12; see also Bernard Grofman & Lisa Handley, Identifying and Remedying Racial Gerrymandering, 8 J.L. & Pol. 345, 365-74 (1992).
. Here it is undisputed that blacks are not spread evenly throughout the lower Shore and that residential patterns there remain significantly segregated by race.
.Political scientists and voting law scholars, however, have proved that any group of voters— regardless of where they live — can be fit into one contiguous district. See Daniel D. Polsby & Robert D. Popper, The Third Criterion: Compactness as a Procedural Safeguard Against Partisan Gerrymandering, 9 Yale L. & Poly.Rev. 301, 331 (1991) ("If every name in the Manhattan phone book is randomly associated with one of ten districts, a map can be constructed that will place every voter in a literally contiguous district no matter which combination of names and districts are chosen. The resulting redistricting map would certainly look odd — in places, districts might be stretched thin as telephone wires — but it can be done, regardless of where the voters live.”). On this view, Justice Brennan’s requirement in Gin-gles that a minority group be compact enough to be placed in a contiguous remedial district would actually be no requirement at all.
. See, e.g., Richard H. Pildes & Richard G. Niemi, Expressive Harms, "Bizarre Districts," and Voting Rights: Evaluating Election-District Appearances After Shaw v. Reno, 92 Mich.L.Rev. 483, 555 & n. 203 (1993) (measuring the ratio of the district's area to the square of the district's perimeter); id. at 557 (measuring the ratio of the district's population to the population of the area that would fall inside a rubber band stretched tightly around the district); Earnest C. Reock, Jr., Measuring Compactness as a Requirement of Legislative Apportionment, 5 Midwest J.Pol.Sci. 70, 71 (1961) (measuring the ratio of the district's area to the area of the minimum circle that could circumscribe the district). See generally Karcher v. Daggett, 462 U.S. at 756 n. 19, 103 S.Ct. at 2673 n. 19 (Stevens, J., concurring) (cataloguing additional measures of compactness).
. Even if District 54-9 were the longest in the State, that might not be fatal. Obviously, longer districts are more acceptable where the population density is low and a considerable area must be covered to maintain equipopulous districts. Other than Garrett County in far western Maryland, Dorchester is the least densely populated county in the State.
. Furthermore, the conclusion that we draw from measuring District 54-9's length and width comports with an "eyeball” assessment: put simply, proposed District 54-9 appears somewhat more spread out and jagged than most of the state plan’s districts, but not significantly so.
. Unlike District 54-9, the NAACP's other proposals included portions of Somerset and (in some instances) Worcester Counties.
. Although the districting criteria set forth in the Maryland Constitution "work in combination ... to ensure the fairness of legislative representation,” it is "a plain fact” that "they tend to conflict in their practical application”; therefore, no redistricting plan can comply to the maximum possible extent with every one of those criteria. In re Legislative Districting, 299 Md. at 681, 475 A.2d at 440; see Legislative Redistricting Cases, 331 Md. at 612-13, 629 A.2d at 665.
. By contrast, the city limits of Salisbury create "islands” of non-city land completely surrounded by city land. It is unclear from the record whether Salisbury’s border — to which the State adhered almost perfectly when drawing the legislative redistricting plan at issue here — was itself an artifact of racial discrimination, as plaintiffs claimed at trial.
. In his dissent Judge Smalkin suggests that the remainder of District 37 which would be left if District 54-9 were removed is not fully contiguous. We read the Cambridge detail map (NAACP D 54, v. 9) as showing that District 54-9 neither touches the Choptank River nor bisects District 37. Furthermore, the southwestern corner of Wicomico County, which the dissent notes would be surrounded on all sides by either water or District 54-9, contains only 164 people and therefore could readily be added to District 54-9. See supra note 34 (discussing the possibility of shifting a few precincts or partial precincts from the remainder of District 37 into District 54-9).
In any event, since plaintiffs have proved their submergence claim, it is the responsibility of defendants to redraw the district lines in a manner that best accommodates the State’s interests while remedying the Voting Rights Act violation. We also note that if the State decides to create a single-member district that is the same (or similar to) proposed District 54-9 and to leave the remainder of District 37 as an at-large, two-member district, the constituents in the remainder will be represented substantially as they would have been under the State’s plan. As a practical matter, Delegates can travel across District 54-9 to meet with constituents in District 37 just as they did before. The critical functional test of compactness clearly can be met. See Dillard v. Baldwin County Bd. of Educ., 686 F.Supp. 1459, 1466 (M.D.Ala.1988).
. Due to an error in the Census Bureau’s computerized map, the NAACP actually incorporated the old, defunct Route 50 bridge. According to testimony at trial, however, the new bridge lies only a few hundred yards northeast of the current bridge and could readily be added to the district, or the Delegate could drive outside the district for a brief moment to meet with constituents on opposite sides of the river.
. As mentioned above, the State then would have to shift a few precincts or partial precincts from District 37 to District 36 or 38 to prevent District 37 from becoming too large. See supra note 34.
. Under the NAACP’s plan, the portion of District 37 not contained in District 54-9 could become a two-member, at-large district (or perhaps two single-member districts). One of the three Delegates currently representing District 37 lives in the proposed single-member District 54-9. The other two incumbent Delegates from District 37 (who live in Caroline and Talbot Counties, respectively) could run in a two-member, at-large race, if they chose to do so. Under the NAACP’s plan, District 38, which currently is represented by one white Senator and three white Delegates, would lose black voters and might gain white voters. See supra note 34 (discussing the possibility of shifting a few majority-white precincts or partial precincts from District 37 to District 38). Therefore, the NAACP’s proposal neither destroys the cores of existing districts nor jeopardizes any advantages that lower Shore residents might derive from being represented in Annapolis by incumbents with seniority-
. Legislative districts have split Wicomico County at least since 1974. See In re Legislative Districting, 271 Md. 320, 332, 317 A.2d 477, 483, cert. denied, 419 U.S. 840, 95 S.Ct. 70, 42 L.Ed.2d 67 (1974).
. Moreover, the Maryland Court of Appeals recently held that a majority-black legislative district "whose creation was mandated by the Voting Rights Act” could properly breach county and municipal boundaries. Legislative Redistricting Cases, 331 Md. at 615, 629 A.2d at 666-67; see also supra note 21 (citing the Supremacy Clause).
. To the extent that District 54-9 does divide municipalities, it does so rather neatly, essentially taking in the southwestern portions of Cambridge, the southern portions of Fruitland, and the western and downtown portions of Salisbury. By contrast, the State’s plan sliced the town of Bladensburg (in Prince George’s County) into three pieces.
. For the last decade, Salisbury has been divided between Districts 37 and 38. See Md. State *1056Gov't Code Ann. § 2-202(1), (37)(f) (1982) (placing all of Wicomico County election district 16— part of which falls within Salisbury's city limits— in District 37). Therefore, the NAACP's proposal to keep Salisbury divided, albeit along somewhat different lines, presents no "new” breach of municipal boundaries.
. Although the State's redistricting plan split 122 precincts, defendants complain that District 54-9 is not compact because it splits 22 of the 32 precincts it touches, thus raising substantial administrative problems. See Burton v. Sheheen, 793 F.Supp. at 1361 (splitting precincts is "primarily an issue which effects administrative convenience”). The Administrator for the State Administrative Board of Election Laws estimated that it would cost about $5,000 per decade for each new precinct created. Even in the worst-case scenario, i.e., that none of the new partial precincts can be absorbed into an adjoining precinct, the creation of the new, majority-black district would cost about $11,000 per year — a small price to ensure fair representation under the law.
. Dr. Lichtman, a renowned authority on the Voting Rights Act and the author of the only statistical textbook devoted exclusively to ecological regression, was the State of Maryland's expert in the present case.. Mr. Hebert is Special Litigation Counsel to the Voting Section of the U.S. Department of Justice’s Civil Rights Division.
. Eighty-seven percent of Dorchester County blacks register as Democrats. Nevertheless, when George Ames, Jr. ran for the Orphans’ Court in 1990 as an independent against three white Democrats, he garnered 93% of the black vote — strong evidence that race predominated over party.
. The black councilman was allowed to "attend[] all meetings except the annual banquet, from which he was excluded. His colleagues sent his dinner on a paper plate to his home.” George H. Callcott, Maryland & America 158 (1985).
. In the litigation that led to the creation of the Dorchester County district, the defendant State of Maryland entered into a consent decree after conceding that the at-large method of nominating and electing county commissioners violated § 2 of the Voting Rights Act.
. The two races were both Democratic primaries: Parren Mitchell's 1986 campaign for Lieutenant Governor and Jesse Jackson’s 1988 campaign for President.
. Furthermore, the Court received testimony about several civil rights groups that have been organizing the Eastern Shore black community across county lines, including targeted efforts to register and turn out voters in Wicomico and Dorchester Counties. That testimony reinforces the direct statistical evidence of black political cohesion throughout District 54-9.
.As quoted above, Growe states: "[T]he 'minority political cohesion’ and 'majority bloc voting' showings are needed to establish that the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population....” - U.S. at -, 113 S.Ct. at 1084.
. In the history of Caroline County, only one African-American has ever won public office: City Commissioner-at-Large James R. Coursey of Denton (which is roughly one-third black). In Talbot County the only African-Americans elected to public office from majority-white areas were two town commissioners in St. Michaels (which also is roughly one-third black), one in the mid-1970s and another in the late 1980s.
. In 1990 Carleton Jones, Sr., an African-American, was elected to the Dorchester County Democratic Central Committee, which is not a public office.
. Other than Judge Barkley and Councilmem-ber Holloway, the only African-American electoral successes were candidates for the Wicomi-co County Democratic Central Committee (again, not a public office), Billy G. Jackson in 1978 and William H. Handy in 1982.
. In Wicomico and Dorchester Counties, black turnout rates (as a percentage of the voting-age population) were approximately half as high as white turnout rates in the 1986 and 1990 general elections.
. In 1988 the General Assembly prohibited dual registration. See Md.Elec.Code Ann. art. 33, § 3-2(d) (1988).
. The Resident Delegate statute is reproduced supra at note 29. Plaintiffs have not brought an independent challenge to the statute, as applied to District 37. Even if they had, we would not need to address the issue because the creation of a majority-black, single-member district in Wi-comico and Dorchester Counties would remove District 37 from the statute’s coverage. See Md. State Gov’t Code Ann. § 2-201(d) (mandating the Resident Delegate scheme for three -member delegate districts only). We do not address whether the statute, as applied to District 38, violates § 2 of the Voting Rights Act. We also do not address whether alternative electoral systems, such as cumulative or limited voting, could remedy minority vote dilution on the Shore.
. In a three-county district, such as District 38, the top vote-getter in the contest for each county seat wins. In a four-county district, such as District 37, the top vote-getter for each county seat is determined, and then, among those four, the top three vote-getters win.
. If the State finds that it is unable to prepare and submit a new redistricting plan by February 28, 1994, the Court will, of course, entertain a motion for a reasonable extension of that deadline.