This case is before us for a second time as a result of a remand by the Supreme Court. Freeman Bosley, Jr., and four other individual plaintiffs (collectively, "the Bosley plaintiffs") appealed the district court's1 grant of summary judgment to the City of St. Louis and various officials (collectively, "the St. Louis defendants") in this Voting Rights Act Sec. 2 challenge to a 1991 reapportionment plan. Because we find that the materials submitted to the district court fail to demonstrate the existence of a genuine issue of material fact as to whether the 1991 reapportionment plan violates Sec. 2, we affirm.2

The city of St. Louis is governed by a Board of Aldermen consisting of twenty-eight aldermen elected from twenty-eight single member wards. Article I, section 3 of the St. Louis City Charter requires the St. Louis Board of Aldermen to redraw the boundaries of the aldermanic wards in accordance with federal decennial census figures. App. at 70.

These census figures reveal the changing composition of the city of St. Louis and the effects of this change upon the composition of the aldermanic wards. According to the 1970 census, used for the 1971 districting scheme, the St. Louis population looked like this:

       % of Total  % of Voting Age      % of Aldermen
       Population    Population     Under 1971 Plan
Black     40.9          35.0                35.7
White     58.7          64.5                64.3
Other      0.5           0.5                 0.0

App. at 33-34, 60, 64-67. This population changed over the ensuing ten years, and by 1980 the census figures which were the basis for the 1981 reapportionment painted a different picture:

       % of Total  % of Voting Age   % of Aldermen
       Population    Population     Under 1981 Plan
Black     45.6          40.3             39.3
White     53.5          58.9             60.7
Other      0.9           0.9              0.0

App. at 32-33, 58, 64-67. The present dispute arose when the Board attempted to follow the mandate to reapportion aldermanic wards in 1991. According to the 1990 census figures, the population of the city of St. Louis breaks down as follows:

       % of Total  % of Voting Age     % of Safe Wards
       Population    Population     Under 1991 Plan
Black     47.5          42.7                42.9
White     50.9          55.8                57.1
Other      1.6           1.5                 0.0

App. at 32, 45, 56.

The population of St. Louis is remarkably segregated. Over 90% of the white population is concentrated in the south and northeastern portions of the city, while the black population is concentrated in the northern and northwestern portions of the city. With the exception of a "corridor" through the central portion of the city, and another corridor along the Mississippi River in the northeast part of the city, at least 75% of the population of almost every St. Louis neighborhood is a single race. Because of the combined effects of geography and segregation, the 1991 aldermanic wards may be drawn to create anywhere from eleven to seventeen safe black wards.

At the time the 1991 redistricting process began, the Board of Aldermen consisted of seventeen white incumbents and eleven black incumbents. After rejecting a plan proposed by Ward 3 Alderman Freeman Bosley, the Board approved a plan which provided for twelve safe black wards. App. at 31-32.

The Bosley plaintiffs, dissatisfied with a failure to obtain fourteen safe black wards, filed a complaint seeking declaratory and injunctive relief, alleging that the St. Louis defendants manipulated ward boundaries by "packing"5 and "cracking"6 black voters for the purpose and with the effect of minimizing the electoral potential of black voters in St. Louis, in violation of Sec. 2 of the Voting Rights Act and the First, Thirteenth, Fourteenth and Fifteenth Amendments. In February, the St. Louis defendants moved for summary judgment on the grounds that the plan's maximum deviation from the ideal ward7 was + 3.98% and - 4.32%, for a total range of 8.27% and a standard deviation of 2.83%. App. at 50. The St. Louis defendants maintained that the wards had been drawn in such a way as to provide substantial proportionality8 for black voters, and that the districting plans had resulted in such proportionality since 1971.

Four affidavits were attached to the St. Louis defendants' motion. Fred Steffen, Clerk of the Board of Aldermen of the City of St. Louis, provided information concerning the identity and race of aldermen who represented each ward for 1971-1992. App. at 63-67. Pelham J. Robinson, Registrar of the City of St. Louis, provided relevant sections of the St. Louis City Charter and a copy of Ordinance 62476, which established the ward boundaries under the 1991 reapportionment plan. App. at 68-83. Stephen Umsheid stated that in a few cases ward boundaries were drawn "through" census blocks, and he explained the method used in such cases. These three affidavits provided the legal and factual background for the statistical analysis contained in the affidavit of Donald L. Davison, the defendants' expert. Davison conducted extensive statistical analyses of the population of St. Louis. Davison analyzed the total population, the voting age population and the Board of Aldermen in terms of racial composition and compared the percentages of each population to the number of aldermen and safe wards. These comparisons led Davison to conclude that the percentage of safe black wards (or black aldermen) was substantially proportional to the percentage of blacks in the voting age population.

The Bosley plaintiffs opposed the motion by filing an affidavit from their own expert, Dr. Charlene Jones. Jones did not take issue with Davison's analysis of proportionality if voting age population is used as the relevant population. Rather, Jones took issue with Davison's choice to use voting age population rather than total population in his analysis of proportionality. Jones stated that blacks have historically been underrepresented in terms of total population, and therefore no proportionality existed. Jones noted the continued effects of past discrimination, a factor under White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), and stated that black voting power had not been maximized. Jones also objected to the Umsheid affidavit, claiming that Umsheid used a flawed methodology and improperly adjusted census figures. Jones further objected to the drawing of ward boundaries so that census blocks were split among two wards.

On May 28, the Bosley plaintiffs moved for leave to file a memorandum and additional affidavits in opposition to the St. Louis defendants' motion for summary judgment. The district court denied the motion for leave to file additional materials and granted the defendants' motion for summary judgment on all plaintiffs' claims. The Bosley plaintiffs filed a Rule 59(e) motion to alter or amend the judgment that again sought to introduce additional factual materials. This motion was denied and the plaintiffs appealed the dismissal of their Voting Rights Act claims, arguing that the district court (1) abused its discretion in denying the motion for leave to file a memorandum in opposition; and (2) erred in granting summary judgment on the Voting Rights Act Sec. 2 claims. On August 4, 1993, we affirmed the district court's summary judgment pursuant to 8th Circuit Rule 47B. African Am. Voting Rights Legal Defense Fund v. Villa, 999 F.2d 1301 (8th Cir.1993) (AAVR ). We denied a petition for rehearing en banc over a dissent by Judge McMillian. Id. at 1302.

On June 30, 1994, the Supreme Court decided Johnson v. De Grandy, --- U.S. ----, 114 S. Ct. 2647, 129 L. Ed. 2d 775 (1994). On the same day, the Supreme Court granted certiorari in AAVR, vacated the panel opinion, and remanded the case for further consideration in light of Johnson. Tyus v. Bosley, --- U.S. ----, 114 S. Ct. 2776, 129 L. Ed. 2d 888 (1994). We have carefully considered the Supreme Court's decision in Johnson and have concluded that the district court's grant of summary judgment was consistent with the principles enunciated in Johnson. Accordingly, we affirm.

The Bosley plaintiffs argue that the district court erred in three respects when it granted summary judgment to the St. Louis defendants. The plaintiffs use Johnson to raise three specific challenges to the district court's analysis of proportionality. First, the Bosley plaintiffs argue that the district court improperly used voting age population rather than total population to define the "relevant population" for assessing proportionality. Second, they claim that the district court erred by including the entire city of St. Louis, rather than only the allegedly challenged areas, when determining the number of majority-minority wards to be considered in measuring proportionality. Finally, they argue that the district court gave excessive weight to the proportionality factor by treating it as a per se bar to a Sec. 2 voting dilution claim. However, before we can address the merits of the Bosley plaintiffs' Voting Rights Act claims, we must decide whether the materials that were untimely filed should have been considered by the district court when determining whether a genuine issue of material fact existed.

A. Did the district court abuse its discretion in refusing to consider the May 28 materials?

The St. Louis defendants filed their motion for summary judgment on February 19, 1992. Local Rule 7(B)(2) provides:

A party opposing a motion for summary judgment under Fed.R.Civ.Pro. 56 shall serve and file any written brief or memorandum of law and any appropriate extra-pleading materials within twenty days after being served with the motion [for summary judgment].

The Bosley plaintiffs moved for an extension of time to respond and filed the Jones affidavit on April 27, the last day of the extension. More than thirty days later after the extension, and more than ninety days after the original twenty days provided in the local rule, the Bosley plaintiffs sought leave to file a memorandum and affidavits in opposition.

The district court has discretion whether to accept or reject such untimely filed materials. Lujan v. National Wildlife Fed'n, 497 U.S. 871, 894-98, 110 S. Ct. 3177, 3191-94, 111 L. Ed. 2d 695 (1990). A district court's refusal to accept untimely filed materials will not be reversed for abuse of discretion unless the proponent of the materials has made an affirmative showing of excusable neglect. Id.

The Bosley plaintiffs submitted a five-paragraph pro se motion to file a memorandum in opposition to the defendants' motion for summary judgment. Citing the importance of the issue, the plaintiffs requested permission to file the memorandum. Although the motion noted the original counsel's failure to file a memorandum or additional affidavits, it was devoid of any reason for this failure to act. Indeed, only by viewing the motion very charitably may any explanation be found. The closest thing to an explanation contained in the motion is the opening paragraph, which states that:

[d]ue to fundamental differences with counsel, we [the Bosley plaintiffs] dismissed our attorneys and advised them of our decision to withdraw from this case so that we could join as plaintiffs in another voting rights case challenging the same St. Louis Ward map brought by another attorney whose strategies we agree with.

App. at 105. This does not explain why the untimely filed materials were not filed on or before April 27; it only explains the reason for the plaintiffs' decision to file the materials on May 28 after dismissing the original attorney. However, assuming that the same strategic disagreements existed during the period leading up to the filing of the Jones affidavit, the district court decision is not an abuse of discretion. The explanation given by the Bosley plaintiffs is an explicit reference to disagreements concerning strategy, and tactical decisions do not amount to affirmative showings of excusable neglect under Rule 6(b).9 See Slaughter v. Southern Talc Co., 919 F.2d 304, 307 (5th Cir.1990) (tactical reasons do not amount to excusable neglect). We believe that the district court did not abuse its discretion when it refused to accept these materials which were submitted in violation of the local rule. Lujan, 497 U.S. at 894-98, 110 S.Ct. at 3191-94 (no abuse of discretion where district court refused affidavits submitted in violation of briefing order and Fed.R.Civ.P. 6(d) and 56(c)); see, e.g., Orsi v. Kirkwood, 999 F.2d 86, 91-92 (4th Cir.1993) (no abuse of discretion where district court refused untimely filed affidavits); Slaughter, 919 F.2d at 307 (no abuse of discretion where district court refused affidavits submitted in violation of Fed.R.Civ.P. 6(d) and 56(c)); Western Chance No. 2 v. KFC Corp., 957 F.2d 1538, 1544 (9th Cir.1992) (no abuse of discretion where district court refused affidavit submitted in violation of local rule); Useden v. Acker, 947 F.2d 1563, 1571 (11th Cir.1991), cert. denied, --- U.S. ----, 113 S. Ct. 2927, 124 L. Ed. 2d 678 (1993) (same). Thus, the district court did not abuse its discretion when it considered the record as consisting of the Jones affidavit and the materials accompanying the St. Louis defendants' motion for summary judgment. Accordingly, we consider the question whether the Jones affidavit and the defendants' materials create a genuine issue of material fact as to whether the 1991 reapportionment plan violated Sec. 2.

B. Did the district court err in granting the St. Louis defendants summary judgment on the basis of sustained proportionality?

Summary judgment is appropriate when there is no disputed issue of material fact and the moving party is entitled to judgment as a matter of law. Egan v. Wells Fargo Alarm Servs., 23 F.3d 1444, 1446 (8th Cir.), cert. denied, --- U.S. ----, 115 S. Ct. 319, 130 L. Ed. 2d 280 (1994); Fed.R.Civ.P. 56(c). We review a grant of summary judgment de novo, applying the same standard as the district court. Egan, 23 F.3d at 1446. Material facts are determined with reference to the substantive law governing the claim.

Section 2 of the Voting Rights Act of 1965, codified as amended, 42 U.S.C. Sec. 1973, provides:

(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title [concerning language minorities], as provided in subsection (b) of this section.

(b) A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office 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 of the population.

42 U.S.C. Sec. 1973.

In Thornburg v. Gingles, 478 U.S. 30, 106 S. Ct. 2752, 92 L. Ed. 2d 25 (1986), the Supreme Court imposed structure upon this totality of the circumstances approach by looking to the Act's legislative history and enunciating a series of factors to be considered. These factors, known as the Gingles factors, include: (1) whether the minority group "is sufficiently large and geographically compact to constitute a majority in a single-member district"; (2) whether the minority group is "politically cohesive"; and (3) whether "the white majority votes sufficiently as a bloc to enable it--in the absence of special circumstances ...--usually to defeat the minority's preferred candidate." Id. at 50-51, 106 S.Ct. at 2766-67.

Although Gingles involved a challenge to a multi-member districting scheme, Voinovich v. Quilter, --- U.S. ----, ---- - ----, 113 S. Ct. 1149, 1157-58, 122 L. Ed. 2d 500 (1993), and Growe v. Emison, --- U.S. ----, ---- - ----, 113 S. Ct. 1075, 1083-84, 122 L. Ed. 2d 388 (1993), make it clear that the Gingles factors are preconditions to a challenge of a single member districting scheme. Johnson reaffirms that the Gingles factors are preconditions in a vote dilution challenge to a single member districting scheme. Johnson, --- U.S. at ----, ----, 114 S. Ct. at 2655, 2657.

In Johnson, the Supreme Court held that substantial proportionality barred a Sec. 2 voting dilution claim where there was a history of discrimination and a two percentage point difference between the percentage of Hispanics in the voting age population (47%) and the percentage of Hispanic-majority districts (45%). The Supreme Court also held that "[w]hile such proportionality is not dispositive in a challenge to single member districting, it is a relevant fact in the totality of circumstances." Id. at ----, 114 S.Ct. at 2651.

Neither party makes any claim that any of the three Gingles preconditions is not satisfied. Rather, as in Johnson, the dispute centers around whether proportionality has been established and, if so, whether proportionality defeats the Bosley plaintiffs' Sec. 2 claim despite the satisfaction of the Gingles factors and the White factor of continuing effects of past discrimination. See id. at ----, ----, 114 S. Ct. at 2655, 2658. The Bosley plaintiffs' claimed errors may be grouped into two categories. The first two arguments challenge the district court's conclusion that proportionality existed. The third argument challenges the district court's treatment of proportionality, assuming that proportionality exists. We address each claimed error in turn.

1. Is there substantial proportionality?

a. Is the total population or the voting age population the relevant population for assessing proportionality?

The parties are in agreement on the material facts. They applied the same criteria for determining whether blacks constituted an effective voting majority in a ward. App. at 87 (Jones affidavit); 32 (Davison affidavit). They agree that it is possible to draw boundaries so as to create fourteen wards with effective black majorities. App. at 89 (Jones affidavit); 22, 23 (defendants' Answer). The Jones affidavit did not take issue with Davison's analysis of proportionality if voting age population is used. Rather, Jones argued that Davison's analysis of proportionality was defective because Davison used voting age population rather than total population. App. at 90.

The Jones affidavit fails to create a genuine issue of material fact as to the existence of proportionality because Jones' premise that total population must be used as the basis for proportionality analysis is incorrect under Johnson. Jones states that blacks have historically been underrepresented in terms of total population. App. at 90-91. The Davison affidavit does not dispute this, and we assume arguendo that blacks have been underrepresented in terms of total population. However, Johnson instructs us to look to voting age population to perform proportionality analysis. Accordingly, we reject the Bosley plaintiffs' claim that the district court erroneously focused on the voting age population rather than the total population when assessing proportionality.

In the first paragraph of Johnson, the Supreme Court framed its holding as:

We hold that no violation of section 2 can be found here, where, in spite of continuing discrimination and racial bloc voting, minority voters form effective voting majorities in a number of districts roughly proportional to the minority voters' respective shares in the voting age population.

--- U.S. at ----, 114 S. Ct. at 2651 (emphasis added). Johnson consistently defines the relevant population as the voting age population. See, e.g., id. at ----, ---- & n. 11, ---- n. 18, ----, 114 S. Ct. at 2654, 2658 & n. 11, 2662 n. 18, 2663. Although footnote 11 of Johnson refuses to resolve the issue of what is the relevant population in a vote dilution claim, the repeated references to voting age population, coupled with the ultimate holding of Johnson, at least permit, even if they do not require, use of the voting age population as a relevant population. Johnson clearly stands for the proposition that voting age population is a relevant population for proportionality purposes despite Johnson 's refusal to hold that it is the only relevant population. Subsequent cases have unanimously interpreted Johnson as approving the use of voting age population as a relevant population. See, e.g., NAACP v. Austin, 857 F. Supp. 560, 569 (E.D.Mich.1994) ("In assessing proportionality we use voting-age population figures, as the Supreme Court did in De Grandy."); see also Shaw v. Hunt, 861 F. Supp. 408, 440 n. 28 (E.D.N.C.), petition for cert. filed, 63 U.S.L.W. 3439 (U.S. Nov. 21, 1994) (No. 94-923); Vera v. Richards, 861 F. Supp. 1304, 1339 n. 49 (S.D.Tex.) (three-judge district court), appeal filed, 63 U.S.L.W. 3388 (U.S. Oct. 31, 1994) (No. 94-805). Accordingly, we hold that the district court's use of voting age population was consistent with Johnson and was not error.10

We hold that under the 1991 plan, proportionality exists between the percentage of blacks in the voting age population (42.7%) and the percentage of safe black wards (42.9%, or 12 out of 28 wards).11 This proportionality has existed since 1971: Under the 1971 plan, the black voting age population was 35.0% of the total voting age population, and 35.7% (or 10 out of the 28 wards) were represented by black aldermen. App. at 33-34. Black representation exceeded black population by 0.7%. Under the 1981 plan, the black voting age population was 40.3% of the total voting age population and percentage of black aldermen was 39.3% (or 11 out of the 28). App. at 32-33. This difference is half of the 2.0% difference approved in Johnson. Johnson, --- U.S. at ----, 114 S. Ct. at 2658 (no denial of equal opportunity where Hispanics were effective voting majority in 45% of districts and constituted 47% of the voting age population). We believe the 1991 plan provides continued proportionality, and that the twenty-year history of proportionality (1971-1991) is of sufficient duration to be considered as "sustained."12 Accordingly, we hold that the district court acted in a manner consistent with Johnson when it examined voting age population to conclude that the St. Louis defendants demonstrated sustained proportionality.

b. Does the proper geographic scope consist of the entire city of St. Louis or the five central corridors?

We also believe that the district court's focus upon the entire city of St. Louis rather than upon the five central corridor wards was consistent with Johnson. The Bosley plaintiffs argue that Johnson requires a focus solely upon the contested areas, which they equate with the five wards. We agree that Johnson stands for the proposition that the proper geographic scope for the comparison is the scope that is pleaded in the complaint and subjected to proof. However, we disagree with the Bosley plaintiffs' revised characterization of their challenge.

In Johnson, which was a consolidation of three actions, "the plaintiffs ... passed up the opportunity to frame their dilution claim in statewide terms." Johnson, --- U.S. at ----, 114 S. Ct. at 2662. The complaints filed by two sets of plaintiffs "pointed to areas around the state" of Florida where the alleged violations occurred. Id. at ----, 114 S.Ct. at 2652. The Department of Justice (DOJ) filed a third action that "alleged that [the apportionment plan] diluted the voting strength of blacks and Hispanics in two parts of the State in violation of Sec. 2."13 Id. "[N]either the evidence at trial nor the opinion of the District Court addressed white bloc voting and political cohesion of minorities statewide." Id. at ----, 114 S.Ct. at 2662. In spite of the fact that the complaints and evidence focused solely upon two counties, the United States sought to expand the geographic scope beyond these counties to include the entire State of Florida. Id. Johnson rejected this attempt, confining the geographical scope of its proportionality analysis to the two counties originally challenged in the complaints and subjected to proof at the five-day bench trial. Id. Johnson concluded its discussion of this issue by noting that the parties "apparently agreed in the District Court on the appropriate geographical scope for analyzing the alleged Sec. 2 violation and devising its remedy." Id.

We believe that the Bosley plaintiffs have fixed upon a citywide geographical scope for analyzing the alleged Sec. 2 violation just as the parties agreed on a two-county-wide scope in Johnson. The Bosley plaintiffs' complaint focuses upon a citywide distribution of voting strength. It alleges a Sec. 2 violation due to the 1991 redistricting plan's failure to provide at least fourteen black majority wards out of a citywide twenty-eight. Compl. pp 45 & 49. Moreover, the complaint refers to the aldermanic wards in a generic fashion that includes all twenty-eight; it does not limit the described wards to the five central corridors. Nowhere in the complaint is any allegation made concerning corridor-specific voting practices; all such allegations are made on a citywide basis. The Bosley plaintiffs' complaint obviously focuses upon the entire city of St. Louis and they have "passed up the opportunity to frame their dilution claim" in more restrictive terms.

In addition, the only proof submitted by the Bosley plaintiffs, the Jones affidavit, discusses citywide voting patterns. Jones states that:

[t]he plan for reapportionment of the twenty-eight wards for election of members to the Board of Aldermen of St. Louis City, Missouri, enacted in December, 1991, diluted the voting strength of black residents in St. Louis City by packing part of the black population into twelve wards ... w[h]ere they constitute an overwhelming black majority in eleven of said wards in excess of 75% and then by cracking the remaining black population among [seven] wards ... in which they constituted a voting age minority of the population....

App. at 86. Jones also states her conclusions as "[t]he 1991 plan for reapportionment of the twenty-eight wards for election of members of the St. Louis City Board of Aldermen and political party committee members has a discriminatory effect," and that "[i]f the present geographical ward boundaries for electing aldermen and political party committee members, were reapportioned fairly and without discriminatory effect, blacks would constitute effective voting majorities in fourteen out of the twenty-eight wards of said city."14 App. at 87. Accordingly, we conclude that the Bosley plaintiffs' pleadings and proof before the district court focused upon a citywide area, not merely the five central corridors. We are not persuaded by the plaintiffs' belated attempts to recharacterize their claims on appeal. Thus, because Johnson defined geographical scope by the scope that was pleaded and the subject of proof, and the Bosley plaintiffs' pleadings and proof concerned the entire city of St. Louis, we hold that the district court's use of the entire city of St. Louis was consistent with Johnson and was proper.15 Accordingly, we hold that the district court correctly determined that proportionality exists under the 1971-1991 plans.

2. Did the district court improperly apply proportionality as an absolute bar to a Sec. 2 dilution claim?

The Gingles factors are not exhaustive. They are necessary, but not sufficient, to establish a Sec. 2 voting dilution violation. Johnson, --- U.S. at ----, 114 S. Ct. at 2657. In Johnson, the Supreme Court considered the existence of proportionality to be sufficient to provide minority voters an equal opportunity to participate in the political process. Id. at ----, 114 S.Ct. at 2651. However, proportionality is only one factor in the totality of circumstances. It is not a "safe harbor" that automatically defeats a claim of vote dilution:

While ... proportionality is not dispositive in a challenge to single-member districting, it is a relevant fact in the totality of circumstances to be analyzed when determining whether members of a minority group have 'less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.'

Id. (citation omitted); see id. at ----, 114 S.Ct. at 2659. The ultimate question is whether the result of manipulation of district lines "interacting 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." Id. at ----, 114 S. Ct. at 2655 (internal citation, quotation and modification omitted).

In its June 17, 1992 order, the district court stated that "[d]efendants contend that as a matter of law Plaintiffs cannot establish their Voting Rights Act claim because evidence of persistent proportional representation establishes that there is no violation of Section 2." Order at 7. The district court then quoted a passage from Gingles that stated that "persistent proportional representation is inconsistent" with a Sec. 2 violation. The district court cited Harvell v. Ladd, 759 F. Supp. 525 (E.D.Ark.1991), rev'd on other grounds, 958 F.2d 226 (8th Cir.1992) (subsequent history omitted), as a case in which sustained electoral success was "the overriding factor." We believe that these quotations and citations indicate a view of the role of the proportionality factor that is consistent with Johnson.

Although, under a strained reading, the district court opinion could be construed as applying a per se rule that proportionality bars a Sec. 2 voting dilution claim, we do not believe that this interpretation is correct.16 Rather, we believe the district court properly considered the proportionality factor as one of several factors. Although the district court's opinion focuses heavily upon proportionality, it addresses the various other factors and points raised by the Jones affidavit. See, e.g., App. at 133-34 (discussing Jones' challenge to Davison's methodology, to the Umsheid affidavit, and to the practice of block splitting).17

However, we need not resolve this issue of interpretation, for assuming that the district court focused its opinion too heavily upon proportionality, we "review judgments, not opinions, and we may affirm a judgment on any ground supported by the record, whether or not that ground was urged below or passed on by the district court." United States v. Sager, 743 F.2d 1261, 1263 n. 4 (8th Cir.1984), cert. denied, 469 U.S. 1217, 105 S. Ct. 1196, 84 L. Ed. 2d 341 (1985); accord United States v. Dugan, 912 F.2d 942, 944 (8th Cir.1990); Johnson, --- U.S. at ---- n. 5, 114 S. Ct. at 2653 n. 5 (discussing review of district court judgment by Supreme Court). We believe that a proper consideration of the proportionality factor as it relates to the totality of circumstances in this case leads inexorably to the conclusion that the complaints contained in the Jones affidavit fail to create a genuine issue of material fact as to whether a Sec. 2 violation has occurred. Jones objected to the drawing of ward boundaries so that census blocks were split among two wards. However, there is no requirement that federal decennial census blocks be used as boundaries during reapportionment. See Burns v. Richardson, 384 U.S. 73, 91, 86 S. Ct. 1286, 1296, 16 L. Ed. 2d 376 (1966). Jones states that black voting power was not maximized. App. at 88, 89. However, black voting power need not be maximized. Johnson, --- U.S. at ----, 114 S. Ct. at 2656. The Jones affidavit recited the Gingles factors and the White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), factor of the continuing effects of past discrimination. App. at 85-86. However, the proportionality factor was given extremely heavy weight in Johnson despite evidence of continuing discrimination and satisfaction of the Gingles and White factors. See Johnson, --- U.S. at ----, 114 S. Ct. at 2663. None of the contents of the Jones affidavit create any dispute that would distinguish this case from Johnson.

It is significant that none of the scenarios that led the Johnson Court to reject a per se rule are present in this case. See id. at ---- - ----, 114 S. Ct. at 2660-61. The only allegation of differential treatment of whites and blacks is Jones' statement that safe black wards contained smaller populations than other wards. To the extent that this is true, it is an enhancement rather than a dilution of the black vote. Moreover, contrary to the assertions of the plaintiffs, the twenty-eight wards were drawn remarkably evenhandedly. They had an average deviation of 2.8% and a range of 8.3%, neither of which is significant. See Brown v. Thomson, 462 U.S. 835, 842, 103 S. Ct. 2690, 2696, 77 L. Ed. 2d 214 (1983) (reapportionment plans with deviations of less than 10% require no justification in Fourteenth Amendment context). We believe that this case is virtually indistinguishable from Johnson, and that a proper consideration of the proportionality factor would yield the same result as that reached by the district court. Accordingly, we hold that the district court did not act in a manner inconsistent with Johnson when it awarded summary judgment to the defendants on the basis of the record before it.18

We hold that the district court did not abuse its discretion when it rejected the Bosley plaintiffs' untimely filed materials. We also hold that the district court properly granted summary judgment where the record before it demonstrated that sustained and substantial proportionality existed between the percentage of blacks in the citywide voting age population and the number of safe black wards. Accordingly, the judgment of the district court is affirmed.