Full opinion text
MEMORANDUM OPINION AND ORDER SHERMAN G. FINESILVER, District Judge: In these consolidated cases, the Court is called upon to select a congressional redistricting plan for the State of Colorado. Following the tabulation of the 1980 decennial census, the Clerk of the United States House of Representatives informed the Governor of Colorado that the state, which currently has five representatives in Congress, was entitled to an additional congressional seat. Although the Governor and the General Assembly (sometimes referred to as the “State Legislature”) made repeated attempts to develop an acceptable redistricting plan, both parties were unable to agree on the composition of the new districts. These suits were filed by several concerned citizens of Colorado in an effort to break the existing stalemate through judicial intervention. After reviewing the testimony and evidence presented at trial, we conclude that the current congressional redistricting plan set forth in C.R.S. 1973 § 2-1-101 is unconstitutional. The Court is of the view, however, that none of the plans submitted to the Court during the course of this litigation fully comport with the objectives and criteria which we feel should be incorporated in a judicially approved redistricting plan. As a result, the Court has fashioned its own plan which satisfies the relevant legal criteria and incorporates the most desirable aspects of the plans presented to the Court. I OVERVIEW OF THE LITIGATION At the outset, we emphasize that “reapportionment is primarily a matter for legislative consideration and determination and that judicial relief becomes appropriate only when a legislature fails to reapportion to federal constitutional requisites in a timely fashion after having had an adequate opportunity to do so.” White v. Weiser, 412 U.S. 783, 794-95, 93 S.Ct. 2348, 2354, 37 L.Ed.2d 335 (1973) (quoting Reynolds v. Sims, 377 U.S. 533,586, 84 S.Ct. 1362, 1394, 12 L.Ed.2d 506 (1964)). The State Legislature fulfilled this responsibility ten years ago when the bill creating the current congressional districts was passed by both houses in the General Assembly and signed into law by the Governor. See C.R.S. 1973 § 2-1-101. At that time, there were 2,209,596 people residing within the borders of this state, and the ideal population of each district was 441,919. Over the past ten years, however, the population of Colorado has grown at a rapid pace. According to the most current 1980 census figures, the population has increased by almost 700,000 people and now stands at 2,889,735. Since this growth rate was proportionally greater than the average overall growth of the nation, Colorado was assigned an additional congressional seat during the decennial reapportionment of the House of Representatives. Under the provisions of both federal and state law, the primary responsibility for drawing new congressional districts lies with the State Legislature, subject to the approval of the Governor. 2 U.S.C. § 2c (1979); Colo.Const. art. V, § 44 (1973). Shortly after the Clerk of the House of Representatives issued the reapportionment mandate, the Governor and the General Assembly began to consider various methods of approaching the task of redistricting. By July 6, 1981, the first proposed plan, designated “H.B. 1615”, had passed both houses of the General Assembly. The Governor promptly vetoed the bill and sent the matter back to the Legislature for further study. His veto message urged the General Assembly to set aside partisan political considerations and develop a “fairer and more responsible plan for congressional redistricting.” Exhibit 27a, Governor’s Veto Message of H.B. 1615 dated June 12, 1981. In response to the Governor’s veto message, the Executive Committee of the State Legislative Council appointed an Interim Committee on Congressional Redistricting to review new proposals and make recommendations to the General Assembly. After several days of meetings in early July, the Committee referred five plans to the General Assembly. Although the Governor notified the Legislature in advance that none of these plans was acceptable, one plan, designated “H.B. 1618” , was subsequently passed by both the State House of Representatives and Senate. The Governor vetoed the bill immediately upon receipt. He criticized the General Assembly for creating a highly partisan plan which would split the home counties of three incumbent Democratic representatives while giving “safe” districts to two incumbent Republicans. In addition to the political problems, the Governor noted that the plan “needlessly [split] select counties and fail[ed] to respect important communities of interest.” He urged the General Assembly to work together toward an acceptable compromise for the citizens of Colorado. Exhibit 27b, Governor’s Veto Message of H.B. 1618 dated July 17, 1981. The matter was sent back to the Legislature’s Interim Committee on Congressional Redistricting which met for several days in August. The Committee reviewed a large number of proposals and ultimately selected three plans for referral to the entire Legislature. Despite warnings from the Governor that he could not approve any of the plans recommended by the Committee, the Legislature reconvened and passed a third plan, designated “H.B. 1624”, on September 22, 1981. Rather than veto the bill immediately, the Governor announced his intention to refrain from taking any action on the measure for ten days to allow for the possibility of a compromise. Before any serious negotiations could begin, however, the first of these consolidated lawsuits, Civil Action No. 81-F-1713, was filed with the Court. The plaintiffs, Judith F. Carstens, Kim M. Rue, W. R. Bray, J. Robert Schafer and Sherill R. Rochford, are residents of each of the five current congressional districts. Their complaint seeks declaratory and injunctive relief against Governor Richard D. Lamm and Secretary of State Mary Estill Buchanan. Specifically, plaintiffs (hereinafter referred to as the “Carstens plaintiffs”) make three requests of this Court: (1) to rule that C.R.S. 1973 § 2-1-101 is unconstitutional because it provides for only five congressional districts instead of the six districts mandated by the 1980 apportionment of the House of Representatives; (2) to enjoin defendants from conducting either the primary or general congressional election for 1982 until Colorado is lawfully divided into six congressional districts; and (3) to accept H.B. 1624, the most recent plan adopted by the Legislature, as the redistrieting plan for Colorado in the event that the Governor and the Legislature could not agree on a compromise proposal in a timely manner. Even the possibility of judicial intervention in the redistricting matter could not prompt the parties to reach an accord. On October 8, 1981, the Governor’s ten-day grace period expired and he vetoed H.B. 1624. For the first time, however, the Governor’s veto message not only included a critique of the rejected plan but also put forth a counter-proposal. Exhibit 27c, Governor’s Veto Message of H.B. 1624 dated October 8, 1981. The Governor subsequently met with leaders of the State Legislature to discuss the merits of his plan and the prospects for resolving the matter without judicial assistance. In the meantime, several concerned citizens filed a second lawsuit on October 23, 1981. The new plaintiffs, David T. Goens, Janet Roberts, Jennie Sanchez, George Rosenberg and Jean Galloway (hereinafter referred to as the “Goens plaintiffs”) are also residents of each of the five current congressional districts. Their complaint similarly requests this Court to enjoin the defendants, Governor Richard D. Lamm and Secretary of State Mary Estill Buchanan from conducting the primary or general congressional elections until a new redistricting plan has been adopted. In the absence of a compromise by the parties, the Goens plaintiffs asked the Court to adopt a plan which satisfied the following seven requirements: (1) population equality; (2) absence of racial discrimination and non-dilution of minority votes; (3) compactness; (4) contiguity; (5) preservation of county lines; (6) preservation of municipal boundaries; and (7) preservation of communities of interest. The Goens plaintiffs also urged the Court to adopt their proposed plan which complied with the seven stated criteria or, in the alternative, to implement provisions of 2 U.S.C. § 2a(c)(2) (1979). Under this federal statute, the state would elect the “additional Representative ... from the state at large and the other Representatives from the districts then prescribed by the law of the state.” 2 U.S.C. § 2a(c)(2) (1979). Negotiations between the Governor and the General Assembly continued throughout the month of October and culminated in a settlement conference at the federal courthouse on November 6, 1981. At the end of this conference, both the Governor and the leaders of the General Assembly conceded that there was no hope of reaching a compromise. The two lawsuits were subsequently consolidated, and the case was set for trial on December 3rd and 4th. In the ensuing weeks, twenty-two plans were submitted to the Court for consideration. At trial, however, testimony focused on five major plans: H.B. 1624, submitted by Carstens plaintiffs; the Goens plan, submitted by the Goens plaintiffs; Governor’s Proposal A and the GRC plan, submitted by defendant Lamm; and the McPhee Plan # 2, submitted by a private citizen. II JURISDICTION The parties agree that this Court has jurisdiction over plaintiffs’ claims for declaratory relief pursuant to 28 U.S.C. § 2201 and 2202 (1979). Jurisdiction over plaintiffs’ equitable claims is based on 28 U.S.C. §§ 1343(3), 1343(4) (1979) and 42 U.S.C. §§ 1983 and 1988 (1979). Because this case challenges the constitutionality of the current apportionment of Colorado’s congressional districts, a three-judge court was convened in accordance with the provisions of 28 U.S.C. § 2284(a) (1979). A. Ripeness Defendant Buchanan contends that the Legislature and the Governor have not had a full opportunity to act on redistricting Colorado, and therefore, the case before the Court is not ripe for adjudication. According to Buchanan, the Legislature and the Governor “may well” reach an accord on redistricting if they are given more time. Plaintiffs, however, point out that Colorado’s congressional districts must be finalized prior to certain precinct and primary deadlines. They argue that the Court cannot wait to take action on this matter until those deadlines expire. Ripeness is an amorphous legal concept subject to many “subtle pressures including the appropriateness of the issues for decision by this Court and the actual hardship to the litigants of denying them the relief sought.” Poe v. Ullman, 367 U.S. 497, 508, 81 S.Ct. 1752, 1758, 6 L.Ed.2d 989 (1961). See also, Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967); In re Grand Jury, 604 F.2d 69 (10th Cir. 1979). The central concern of a ripeness inquiry is “whether there is a substantial controversy, between parties having adverse legal interests, of sufficient immediacy and reality” to warrant the attention of the Court. Lake Carriers Ass’n. v. MacMullan, 406 U.S. 498, 506, 92 S.Ct. 1749, 1755, 32 L.Ed.2d 257 (1972) (quoting Maryland Cas. Co. v. Pacific Coal & Oil Co., 312 U.S. 270, 273, 61 S.Ct. 510, 512,85 L.Ed. 826 (1941)). As a general rule, the Court will not review matters involving uncertain and contingent future events that may not occur as anticipated or may not occur at all. See Wright & Miller, 13 Federal Practice & Procedure, § 3532, p. 238 (1975). In the instant ease, the Governor and the State Legislature have had ample opportunity to redistrict Colorado to provide for an additional representative. The Legislature has passed three different redistricting proposals, all of which were subsequently vetoed by Governor Lamm. No attempt was made to override these vetoes. Despite months of deliberate study and negotiations, the people of Colorado still do not have an acceptable congressional redistricting plan. We are persuaded that the fate of redistrieting in Colorado has reached an impasse which the parties are not capable of resolving. At the end of the November 6 negotiating session in the federal courthouse, the parties agreed that it “would do neither side any ... good to negotiate . .. further”. Transcript, Hearing of November 6, 1981, p. 3. Moreover, at trial Representative Ronald Strahle, Majority Leader of the Colorado House of Representatives testified that it was “highly remote” that the Governor and the Legislature would ever agree on an acceptable redistricting plan. This assessment was never challenged. The hardship to the parties and to the people of Colorado is evident. There is no dispute that the current posture of Colorado’s congressional districts is unconstitutional. If an acceptable redistricting plan is not adopted, the only alternative would be to hold the 1982 congressional elections pursuant to district lines which are patently offensive to the long-established principle of “one person-one vote”. Thus, this Court is presented with a controversy of immediate concern. There is no indication that absent judicial intervention, a viable solution will be forthcoming. Considering the nature of the issues presented for decision and the hardship to the parties, we find that the redistricting question is ripe for adjudication. B. Applicability of 2 U.S.C. § 2a(c)(2) One possible short term solution to the redistricting dilemma is contained in 2 U.S.C. § 2a(c)(2) (1979). This federal statute declares, “until a state is redistricted in the manner provided by the law thereof after any apportionment, ... if there is an increase in the number of representatives, such additional representative or representatives shall be elected from the state at large and the other representatives from the districts then prescribed by the law of such State.” Defendant Buchanan maintains that even if this matter is ripe for review, this Court must defer to the directives of this statute and dismiss the case. The Carstens plaintiffs claim that this Court is not constrained by the provisions of 2 U.S.C. § 2a(c)(2) because the measure was impliedly repealed when Congress passed 2 U.S.C. § 2c in 1967. Section 2c provides, “In each state entitled ... to more than one representative under an apportionment, . . . there shall be established by law a number of districts equal to the number of representatives to which such state is entitled, and Representatives shall be elected only from districts so established, no district to elect more than one representative ...” On its face, Section 2c appears to prohibit at-large elections under any circumstances since the number of districts and the number of representatives must always be equal. We are of the view that these two statutes are not necessarily inconsistent and do not preclude this Court from considering the redistricting dilemma. Section 2c prohibits a legislature or court from deliberately designing a redistricting plan which would elect at-large representatives. Arguably, Congress did not repeal Section 2a(c)(2) because they did not want to leave a state without a remedy in the event that no constitutional redistricting plan exists on the eve of a congressional election, and there is not enough time for either the Legislature or the courts to develop an acceptable plan. In this very limited circumstance, we believe that Section 2a(c)(2) provides emergency statutory relief from an otherwise unconstitutional situation. There is nothing in the language of Section 2a(c)(2) which indicates that Congress intended to bar the federal courts from providing timely assistance to the state in resolving a redistricting dispute. Since the posture of this case does not require the use of such a drastic emergency remedy, we find that Section 2a(c)(2) is not applicable here. Ill ISSUES The issues in this case can be divided into two broad categories. The first category defines the scope of the Court’s authority to act while the second category delineates the criteria which form the basis of the Court’s decision. A. The Scope of the Court’s Authority to Act Although we have established that this Court has jurisdiction to rule on the redistricting dispute, it is also important to determine the scope of the Court’s authority in this matter. According to the Carstens plaintiffs, our authority is somewhat limited. Since redistricting is primarily the responsibility of the State Legislature, they argue that H.B. 1624, the last legislative pronouncement, represents current state policy on redistricting and should receive priority during the Court’s deliberations. Carstens plaintiffs claim that courts in general have paid great deference to the “studied and thoughtful approach” to redistricting provided by the legislative process. As examples of instances in which a court has deferred to the legislature, they cite White v. Weiser, 412 U.S. 783, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973) and Donnelly v. Meskill, 345 F.Supp. 962 (D.Conn.1972). These cases fail to support Carstens’ position. In White v. Weiser, 412 U.S. 783, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973), the Supreme Court reversed a three-judge district panel’s selection of a redistricting plan because that plan “broadly brush[ed] aside state apportionment policy without solid constitutional or equitable grounds for doing so.” Id. at 796, 93 S.Ct. at 2355. The case is not controlling, however, because the “state apportionment policy” referred to by the Court was embodied in a plan which not only had been passed by the State Legislature but also had been signed into law by the Governor. Id. at 784, 93 S.Ct. at 2349. In the instant case, H.B. 1624 was approved by the General Assembly but vetoed by the Governor. Although Donnelly v. Meskill, 345 F.Supp. 962 (D.Conn.1972), another case cited by the Carstens plaintiffs, is factually similar to the situation before this Court, Donnelly is also distinguishable. In that case, the Court was required to select one of three plans submitted after the Legislature’s redistricting bill had been vetoed by the Governor. The plan chosen by the Court contained districts “essentially as outlined by the legislature, with adjustments necessary to bring about virtually complete population equality.” Id. at 965. While the Court felt that this essentially legislative plan was constitutionally sound and workable, it concluded, “if time permitted extended hearings before the court or extended consideration by a court-appointed master, a better plan might be devised, weighing all possible factors.” Id. at 965. In the instant case, the Court has solicited extensive submissions from the parties and does not face the same severe time constraints which confronted the Donnelly court. Thus, we do not feel that the holding in either White or Donnelly compels us to give priority to H.B. 1624, particularly if a better plan is available. Congressional redistricting is a law-making function subject to the state’s constitutional procedures. Koenig v. Flynn, 285 U.S. 375, 52 S.Ct. 403, 76 L.Ed. 805 (1932). See also Smiley v. Holm, 285 U.S. 355, 52 S.Ct. 397, 76 L.Ed. 795 (1932). The Colorado Constitution explicitly provides that every bill passed by the General Assembly shall be signed by the Governor before it becomes law. Colo.Const. art. IV, § 11. There is no authority conferred upon the General Assembly “to create congressional districts independently of the participation of the Governor as required by the state constitution with respect to the enactment of laws.” Smiley, 285 U.S. at 373, 52 S.Ct. at 401. Accord, State ex rel. Reynolds v. Zimmerman, 22 Wis.2d 544, 126 N.W.2d 551 (1964). As a result, H.B. 1624, which certainly is entitled to careful consideration by this Court, cannot represent current state policy any more than the Governor’s proposal. Both the Governor and the General Assembly are integral and indispensable parts of the legislative process. To take the Carstens’ position to its logical conclusion, a partisan state legislature could simply pass any bill it wanted, wait for a gubernatorial veto, file suit on the issue and have the Court defer to their proposal. This Court will not override the Governor’s veto when the General Assembly did not do so. Instead, we regard the plans submitted by both the Legislature and the Governor as “proffered current policy” rather than clear expressions of state policy and will review them in that light. B. The Criteria The primary goal of an acceptable congressional redistricting plan should be “fair and effective representation of all citizens.” Reynolds v. Sims, 377 U.S. 533, 565-66, 84 S.Ct. 1362, 1383, 12 L.Ed.2d 506 (1964). Prior to the early 1960’s, the Supreme Court refused to get involved in the predominantly political cases which challenged the composition of legislative or congressional districts. See e.g., Colgrove v. Green, 328 U.S. 549, 66 S.Ct. 1198, 90 L.Ed. 1432 (1946). The disparities in the composition of districts in some states became so great, however, that the Court was forced to intervene in order to protect the integrity of the legislative system. Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962). Development of the criteria which currently govern congressional redistricting began shortly thereafter with Wesberry v. Sanders, 376 U.S. 1, 84 S.Ct. 526,11 L.Ed.2d 481 (1964). In Wesberry, several citizens challenged the gross population disparities in Georgia’s congressional districts which ranged in size from 823,680 residents in the largest district to 272,154 residents in the smallest district. These individuals claimed that their right to vote had been diluted by the failure of the Georgia Legislature to realign each district to equalize the population. The Supreme Court agreed and established population equality as the first relevant constitutional standard for congressional redistricting. The Court held, “the command of Article I, Section 2 that Representatives be chosen ‘by the People of the several States’ means that as nearly as is practicable one man’s vote in a congressional election is to be worth as much as anothers.” Id. at 7-8. Over the next few years, the Court attempted to define the acceptable parameters of population deviation under the “one person, one vote” principle. Although the Court seemed to assume that equal representation was synonymous with equal population, it steadfastly refused to quantify the criteria announced in Wesberry as a fixed or numerical percentage. In Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519 (1969), the Court invalidated a Missouri congressional redistricting plan with an overall population deviation of 5.97%. Justice Brennan declared that Wesberrys “as nearly as practical” standard required “the State [to] make a good faith effort to achieve precise mathematical equality” and “[to] justify each variance, no matter how small.” Id. at 530-31, 89 S.Ct. at 1228. The Court concluded that the population variances among the Missouri congressional districts were unacceptable and could not be justified by the state’s attempt to avoid fragmenting political subdivisions or areas with distinct economic and social interests, to recognize projected population shifts, to ensure geographically compact districts, or to achieve a reasonable legislative compromise. Id. at 533-36, 89 S.Ct. at 1230. Similarly, in Wells v. Rockefeller, 394 U.S. 542, 89 S.Ct. 1234, 22 L.Ed.2d 535 (1969), the Court invalidated a New York congressional redistricting plan with a total population deviation of 13.095% In that case, New York did not claim that the Legislature had made a good faith effort to achieve mathematical equality. Instead, the state attempted to justify existing deviations by creating districts with specific interest orientations. The Court rejected this argument as “antithetical to the basic premise of the constitutional command to provide equal representation for equal numbers of people.” Id. at 546, 89 S.Ct. at 1237. Although the Court has accepted greater population deviations for state legislative districts, it seems preoccupied with the notion of precise mathematical equality for congressional districts. The most recent decisión on this issue not only affirmed the strict standard enunciated in Kirkpatrick and Wells, but also tightened the range of acceptable variances. In White v. Weiser, 412 U.S. 783, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973), the Court rejected a Texas redistricting plan with an overall population variance of 4.13% in favor of a plan with a total deviation of 0.149%. The current status of the population equality standard is perhaps best described as entrenched ambiguity. No one will deny that it is the “pre-eminent, if not the sole, criterion on which to adjudge [the] constitutionality” of congressional redistricting plans. Chapman v. Meier, 420 U.S. 1, 23,95 S.Ct. 751, 764, 42 L.Ed.2d 766 (1975). The precise point at which a plan passes into the range of acceptable population deviation remains undefined and presumably varies with the plan. In applying this standard, however, the district courts tend to invalidate deviations which are greater than those found in Kirkpatrick and Wells. See e.g., David v. Cahill, 342 F.Supp. 463 (D.C. N. J.1972) (total population deviation of 12.41%); Skolnick v. Illinois State Electoral Board, 307 F.Supp. 698 (N.D.Ill.1969) (total population deviation of approximately 13.6%). Conversely, plans with lower population variances have been approved. See e.g., Dunnell v. Austin, 344 F.Supp. 210 (E.D.Mich.1972) (total population deviation of .01005%); Drum v. Scott, 337 F.Supp. 588 (M.D.N.C.1972) (total population deviation of 3.79%); West Virginia Civil Liberties Union v. Rockefeller, 336 F.Supp. 395 (S.D. W.Va.1972) (total population deviation of O. 07865%); Skolnick v. State Electoral Board of Illinois, 336 F.Supp. 839 (N.D.Ill. 1971) (total population deviation of 1.144%). The second constitutional criteria used in analyzing redistricting plans developed out of the realization that sheer mathematical equality may not adequately protect minority rights. Although no group has a constitutional right to be represented in a legislative body in direct proportion to its numerical voting strength, the Court began to recognize that district lines could be drawn in a manner which would “minimize or cancel out the voting strength of racial or political elements of the .. . population.” Fortson v. Dorsey, 379 U.S. 433, 439, 85 S.Ct. 498, 501,13 L.Ed.2d 401 (1965). See also Whitcomb v. Chavis, 403 U.S. 124, 91 S.Ct. 1858, 29 L.Ed.2d 363 (1971). Thus, plans which satisfy equal population requirements may still be vulnerable to attack on the issue of invidious racial discrimination. Unlike population equality which is capable of quantifiable verification, dilution of minority voting strength addresses the quality of representation and cannot be established through simple numerical tests. Whitcomb, 403 U.S. at 142, 91 S.Ct. at 1868 (1971). Instead, the court must determine whether there is sufficient “evidence to support findings that the political processes leading to nomination and election were not equally open to participation by the group in question — that its members had less opportunity than did residents in the district to participate in the political processes and to elect legislators of their choice.” White v. Regester, 412 U.S. 755, 765-66, 93 S.Ct. 2332, 2339, 37 L.Ed.2d 314 (1973). The constitutionally protected right is one of equal access to the political process. In the instant case, the parties have not challenged an existing redistricting plan on the basis of invidious racial discrimination. Therefore, we find it unnecessary to explore the nuances of proof needed to support such a claim. We are, however, acutely aware of the potential for minimizing the voice of minority populations in the development of redistricting plans and have carefully reviewed all of the proposals submitted to the court with this issue in mind. In our view, a redistricting plan which satisfies this criteria should not fracture a natural racial or ethnic community or otherwise dilute minority voting strength. These two constitutional criteria, population equality and absence of racial discrimination, have formed the foundation of judicial analysis on this issue for almost two decades. While the integrity of these standards has never been seriously challenged, the lower courts have frequently been forced to resort to additional non-constitutional criteria in their comparisons of proposed congressional plans. The degree of sophistication in redistricting technology has reached a point where it can match the courts’ increasing demands for mathematical precision. As a result, courts are often faced with situations in which several different redistricting plans achieve virtually identical levels of population equality without substantially diluting minority rights. In these cases, no reasoned decision can be based solely on these two constitutional criteria. The court must accommodate other relevant criteria in determining whether to accept a proposed plan or to adopt a new one. Therefore, we decided to evaluate the plans submitted to the Court on the basis of several additional non-constitutional criteria. These criteria can be grouped into three categories: 1) compactness and contiguity; 2) preservation of county and municipal boundaries, and 3) preservation of communities of interest. Selection of these criteria was based primarily upon their successful use in other redistricting cases. See e.g., David v. Cahill, 342 F.Supp. 463 (D.C.N.J.1972); Preisler v. Secretary of State of Missouri, 341 F.Supp. 1158 (W.D. Mo.1972); Skolnick v. State Electoral Board of Illinois, 336 F.Supp. 839 (N.D.Ill.1971). We note, as well, that both the Governor and the State Legislature’s Interim Committee on Congressional Redistricting considered these factors important enough to include them on lists of guidelines to be used in developing proposed plans. In addition, the people of Colorado have adopted an amendment to the state constitution which endorses the use of similar criteria in state reapportionment cases. Throughout this litigation, Carstens plaintiffs have criticized the use of non-constitutional criteria in analyzing redistricting plans. They argue that there are internal conflicts inherent in such criteria which make their use subjective and impractical. We do not find these arguments persuasive. When faced with potentially conflicting interests, courts generally attempt to achieve an equitable result by using a balancing test. See David v. Cahill, 342 F.Supp. 463 (D.C.N.J.1972); Skolnick v. State Electoral Board of Illinois, 336 F.Supp. 839 (N.D.Ill. 1971). If conflicts do, in fact, occur when these criteria are applied to redistricting plans, we are convinced that balancing the competing interests represents a realistic and practical means for resolving any potential problems. IV GEOGRAPHIC AND DEMOGRAPHIC PROFILE OF COLORADO Before proceeding to critique the plans submitted to the Court under the criteria outlined above, it is important to review the topography and population distribution of the state. An understanding of geography and demographics is important to the development of a rational state policy for redistricting Colorado. Geographically speaking, the state is divided into three principal regions: (1) eastern plains, (2) western slope and (3) Rocky Mountains and Continental Divide. The eastern portion of Colorado (generally referred to as the eastern plains) covers more than one-third of the state’s land area. This region has flat plains and broad rolling prairies which gradually rise to the foothills and the mountain ranges (on the west) that divide the state. A combination of physical and economic geography influences the eastern portion of the state. These high plains are bisected by two prominent river valleys, the Arkansas and South Platte, which have been dominant forces in the development of this region. The people in the area are dependent on these two rivers for their water supply which is their economic base. The western third of Colorado (commonly referred to as the western slope) consists of alpine terrain interspersed with wide valleys, rugged canyons, high plateaus and deep basins. Like the eastern plains, the west slope contains two prominent river systems, the Colorado and the Gunnison, which converge at the city of Grand Junction, the major population center of the western slope. The Rocky Mountains cut through the middle third of Colorado from north to south and also rise in the northwestern corner of the state. Fifty-five peaks which tower 14,000 feet or more above sea level are found in this mountainous area. The Continental Divide, a line of mountain summits which separates streams flowing toward the Pacific Ocean and the Gulf of California from those flowing toward the Gulf of Mexico, traverses the west central part of the state. This world famous Divide is crossed by a limited number of highways which are open year round. As a result, the Divide has a substantial effect on the distribution of Colorado’s population. The geographic location and physical characteristics of the Divide must therefore be a realistic consideration in any redistricting plan. The distribution of the state’s population over Colorado’s broad and varied geographic spectrum is truly unique. Nine of the state’s 63 counties contain approximately 80 percent of the state’s total population. These nine counties (Adams, Arapahoe, Boulder, Denver, El Paso, Jefferson, Larimer, Pueblo and Weld) are located along the narrow strip of the eastern front range of the Rocky Mountains. The other 20 percent of the state’s population is distributed over a vast geographic area. The retail trade centers, educational institutions, and recreational facilities of the front range counties, however, act as the lifeblood for a major portion of the state. With these facts in mind, we will now examine the plans submitted to the Court under the criteria we outlined above. V CRITIQUE OF THE PLANS UNDER COURT-ADOPTED CRITERIA A. The Constitutional Criteria The first constitutional criteria, population equality, has not been a topic of dispute during the course of this litigation. All of the plans submitted to the Court made a “good faith effort” to comply with this requirement. See Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519 (1969); Wells v. Rockefeller, 394 U.S. 542, 89 S.Ct. 1234, 22 L.Ed.2d 535 (1969). If the 0.149% total deviation figure accepted by the Supreme Court in White v. Weiser, 412 U.S. 783, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973), is used as a benchmark, each of the plans fall well within constitutional limits. The overall population deviation figures range from a low of 7 people or .0015% on H.B. 1624 to a high of 15 people or .0031% on the Governor’s Plan. In simple terms, these figures mean that the drafters of H.B. 1624, the most exacting plan in population terms, came within fifteen one-thousandths of a percent of being able to split the state into six equal congressional districts. The least successful plan, however, was off by only thirty-one one-thousandths of a percent. To select one plan over another on the basis of population equality when only sixteen one-thousandths of a percent separates the five major plans presented to the Court ignores the realities of “fair and effective representation.” We refuse to relegate congressional redistrieting to a hair splitting game in which the citizens of the state emerge as the only real losers. We must therefore rely on the other criteria to draw some important distinctions between these plans. Each of the plans submitted to the Court was carefully analyzed on the basis of the second constitutional criteria, the absence of racial discrimination and the non-dilution of minority votes. The State of Colorado has a combined minority population of 16.9% which is concentrated in three major areas. Exhibit 33(a). The City and County of Denver supports both a substantial Black and Hispanic population. Large groups of Hispanics also reside in the City of Pueblo, elsewhere in Pueblo County and in the San Luis Valley located in the south central portion of the state. Exhibits 13, 14 and 32(a). The manner in which each of the proposed plans deals with these centers of minority population is an important factor in determining their constitutionality and credibility. None of the plans shows any evidence of invidious racial discrimination. At least one, however, raises questions regarding the dilution of minority votes. H.B. 1624, submitted by the Carstens plaintiffs, includes a significant split of the minority populations in both Denver and Pueblo. This plan divides the Denver community in half and likewise splits Pueblo along an irregular configuration. Although Carstens plaintiffs admit that the Denver metropolitan area represents the only concentration of minority strength sufficient to create a relatively strong minority district, they claim that H.B. 1624 splits Denver in order to “enhance” minority votes. This alleged enhancement is accomplished by adding the growing minority population in western Adams county to the large minority population residing in northeast Denver. Thus, Carstens plaintiffs claim that District 1 of H.B. 1624 encompasses substantially all of the expected minority population growth in the Denver metro area in the coming decade. They strongly suggest that “the affirmative goal of having a district that will be one in which a Black-Hispanic minority finally achieves such strength as to guarantee representation of their choice is plainly obtainable under H.B. 1624.” Carstens Closing Argument at p. 5. We have thoroughly examined the minority population statistics for the Denver area and have carefully reviewed the testimony at trial. On the basis of this evidence, we are not persuaded by the arguments of the Carstens plaintiffs. No one will deny that the enhancement of minority voting strength is a worthy goal. The experts at trial agreed, however, that the benefit obtained through a 3 to 5% increase in the minority population in H.B. 1624’s first congressional district was far outweighed by the detrimental impact of splitting the City and County of Denver into several districts. As a general rule, minority voting strength is impermissibly diluted when large concentrations of minority population are necessarily fragmented and disbursed. Beer v. United States, 425 U.S. 130, 141, 96 S.Ct. 1357, 1363, 47 L.Ed.2d 629 (1976); Mississippi v. United States, 490 P.Supp. 569, 581 (D.C.D.C.1979). H.B. 1624 severs traditional Hispanic communities in west Denver, placing part of them in the first district and part of them in the sixth district. The split is made to achieve a 3 to 4% increase in the minority population in H.B. 1624’s first congressional district relative to the other plans. If this increase made a meaningful impact on minority voting strength, we might be more receptive to the position of the Carstens plaintiffs. But the fact remains that a minority population must have a majority of 60 to 65% of the voters in a district in order to exercise political control over that district. United Jewish Organizations v. Carey, 430 U.S. 144, 97 S.Ct. 996, 51 L.Ed.2d 229 (1977); Beer v. United States, 425 U.S. 130, 96 S.Ct. 1357, 47 L.Ed.2d 629 (1976). Unfortunately, it is mathematically impossible to design a district in Colorado in which minorities are “guaranteed representation of their choice.” Cf. Carstens’ Closing Argument at p. 5. Even allowing for substantial growth in the minority population during the next decade, it is unlikely that the minority population in this district will even begin to approach the 60-65% goal. Drawing district lines to enhance minority representation thus becomes a hollow concept, particularly when the 3-4% enhancement in one district is accomplished at the cost of 5-6% reduction in another. Carstens plaintiffs contend that the split through the city and county of Pueblo similarly enhances minority voting strength in the third district. The evidence submitted at trial, however, belies this claim. The district lines which wind and twist through southern Colorado succeed in consolidating the Hispanic population of western Pueblo County and the San Luis Valley in District 3 while placing the Hispanics in Las Animas County in District 4 and the Hispanics in the eastern half of Pueblo County in District 5. As a result, H.B. 1624 fractures the southern Hispanic community more than any other plan before this Court. Thus, after examining the second constitutional criteria, we are no closer to selecting a plan which provides fair and effective representation for the people of Colorado than we were before. While H.B. 1624 has the lowest population deviation, it is the least desirable in terms of minority representation. At least two of the other plans, the Governor’s Plan and Goens Plan, also make minor splits in Denver which affect traditional minority communities. While these splits are probably not sufficient to dilute minority voting strength, they detract from the overall acceptability of these plans. Since we are unable to make meaningful distinctions between the five major plans on the basis of constitutional standards, we will examine them in light of our non-constitutional criteria. B. The Non-Constitutional Criteria Compactness and contiguity represent the first and probably the least significant of the three non-constitutional criteria which we have adopted. Since both of these concepts focus primarily on the geographic shape of the proposed districts rather than on substantive aspects of representation, there is more flexibility in their application. Both were originally designed to represent a restraint on partisan gerrymandering and should be used with this thought in mind. The compactness requirement specifies that the boundaries of each congressional district shall be as short as possible. Although there are several ways to measure compactness, one of the most accurate is “to determine the smallest circle into which the district can be circumscribed and to compare the ratio of the area of the district inside the circle to the area of the circle itself.” American Bar Association Special Committee on Election Law and Voter Participation, Congressional Redistricting 13 (1981) [hereinafter referred to as “Congressional Redistricting”]. The closer these figures come to a 1 to 1 ratio, the more compact the district will be. In a practical sense, the compactness of a congressional district will be directly affected by the density and distribution of a state’s population. Since population requirements have priority, compactness must often be sacrificed in order to achieve an acceptable range of population deviation. Compact districts do, however, reduce electoral costs (in both time and money) and increase the opportunities for more effective representation by concentrating a congressperson’s constituency in an easily accessible area. Congressional Redistricting at 13. The courts which have utilized the compactness criteria have almost always considered contiguity as well. See e.g., Dunnell v. Austin, 344 F.Supp. 210 (E.D. Mich.1972); Preisler v. Secretary of State of Missouri, 341 F.Supp. 1158 (W.D.Mo. 1972). This factor specifies that “no part of one district be completely separated from any other part of the same district.” Dixon, Fair Criteria and Procedures for Establishing Legislative Districts, 9 Policy Studies Journal 839, 847 (Special Issue # 3, 1980-81) [hereinafter referred to as “Criteria and Procedures”]. The universal acceptance of the need for contiguous congressional districts indicates the pragmatic character of this requirement. None of the plans submitted to the Court can be seriously challenged on either of these criteria. The demographics of the State of Colorado are such that the Denver metropolitan districts will be extremely compact while the west slope and eastern plains districts will be quite large. The geographic size of these larger districts cannot be significantly reduced due to the sparse population of these areas. Since each of the plans contained six properly contiguous districts, the first category of non-constitutional criteria has not been helpful in distinguishing among the proposed plans. The next category focuses on the preservation of county and municipal boundaries. These political subdivisions should remain undivided whenever possible because the sense of community derived from established governmental units tends to foster effective representation. See Dunnell v. Austin, 344 F.Supp. 210 (E.D. Mich.1972); Skolnick v. State Electoral Board of Illinois, 336 F.Supp. 839 (1971); Maryland Citizens Committee for Fair Congressional Redistricting, Inc. v. Tawes, 253 F.Supp. 731, aff’d. sub nom., Alton v. Tawes, 384 U.S. 315, 86 S.Ct. 1590, 16 L.Ed.2d 586 (1966). Unnecessary fragmentation of these units not only “undermines the ability of constituencies to organize effectively but also ... increases the likelihood of voter confusion regarding other elections based on political subdivision geographies.” Congressional Redistricting at 12. The priority given to population equality makes the division of some county and municipal lines unavoidable. It is less certain, however, when faced with the choice of preserving county or municipal boundaries, which of these boundaries should prevail. As a general rule, county lines are more meaningful in sparsely populated areas because the residents rely on the county government to provide all necessary services. Municipal boundaries, on the other hand, take precedence in densely populated areas. These local units of government represent logical centers of community interest for urban residents who identify more closely with municipal rather than county services. See David v. Cahill, 342 F.Supp. 463, 469 (D.C.N.J.1972). The State of Colorado is divided into 63 counties which range in population from 408 to almost 500,000. Exhibits 21 and 21A. It is inevitable that some county and municipal boundaries will be split when congressional districting lines are drawn. We believe, however, that any splits in city and county boundaries should be made in a rational manner which attempts to minimize divisions in these local governmental units. The two areas of prime concern in this regard are Denver and Pueblo. The current population of the City and County of Denver is 492,365. Exhibit 21A. Since the ideal population of a new congressional district is 481,622.5, approximately 3% of Denver’s population must be severed from the city in order to achieve population equality. We are of the view that few, if any, of the plans submitted to the court split Denver in a desirable location. The Goens Plan makes this “required” split in the northwest corner of the city, while the Governor’s Plan removes an area from the north central section of Denver. Evidence at trial indicated that both of these splits were undesirable because they separated a significant minority population within Denver. For this reason, we conclude that neither plan embodies the most rational division of the City and County of Denver. H.B. 1624 and the McPhee Plan # 2 split Denver for the alleged purpose of. enhancing minority votes in the first congressional district. As we noted earlier, the 3 to 4% increase in minority population is not enough to have a significant impact on the election. In addition, there was substantial evidence presented at trial which indicated that bisecting the City and County of Denver was not conducive to fair and effective representation. State Senator Ted Strickland, Chairman of the Legislature’s Interim Committee on Congressional Redistricting, testified that the unique nature of Denver’s problems presented a “compelling reason” for maintaining the city as one congressional district. Denver City Council President William Roberts elaborated on these problems when he explained the very different philosophies held by Denver and its suburbs with regard to housing, schools, employment and social services. This point was also reiterated in the affidavit of Cathy Reynolds, Denver City Councilwoman-at-large and representative of the National League of Cities. She emphasized that the division of loyalties which would result from splitting Denver into two districts would only “help perpetuate an already existing urban-suburban split and ... weaken Denver’s strength.... [The city’s] governmental structure, ... taxing ability, ... location and . .. services not only provide' internal bonds within [its] citizenry, but [also] make it practical and logical for [Denver] to have one strong voice in the U. S. House of Representatives.” Exhibit 38B. As a result, we believe that the division of Denver which appears in H.B. 1624 and McPhee Plan # 2 lacks vision and an understanding of the objectives of effective and accountable congressional redistricting. One of the most controversial issues at trial involved the status of Pueblo County in the new redistricting plan. Considerable evidence was presented on the importance of keeping both the city and the county of Pueblo intact. See Exhibits 38d and 42. The only two plans which split this area are H.B. 1624 and McPhee Plan # 2. H.B. 1624 divides the city and the county in half along a “subterranean geological formation” known as the hogback. The McPhee Plan, on the other hand, splits Pueblo in the southwest corner of the county. Carstens plaintiffs maintain that these “sacrifices” were necessary to prevent Pueblo from dominating the western slope district. We do not believe that a reasonably fair and effective congressional redistricting plan should “sacrifice” any county in the state. Pueblo’s 125,972 citizens represent approximately one-fourth of the population needed to create a congressional district. If the county is kept intact, Pueblo will account for only 26% of the population in any district in which it is located. The remaining counties would therefore have a 3 to 1 advantage over Pueblo in population and could effectively counter any concerted effort by Pueblo to unduly influence representation of the area. Every major plan submitted to the Court splits at least five counties. Although most of these splits are necessary to achieve population equality, they are frequently made without regard to the impact such a split would have on the representation of the divided communities. For example, the GCR plan splits Boulder County by placing the northeast corner in a predominantly agricultural district. The City of Boulder, however, is playing a central role in the development of this entire area. As a result, we feel that the needs of these people are better served by a united Boulder County- Similarly, the GCR Plan makes an obscure split in Las Animas County (located in Southern Colorado). This split does not appear to reflect any particular geographic, demographic or economic consideration. In such a sparsely populated, predominantly rural community where the citizens depend upon the county for important services, an unexplained split of even a minimal number of citizens does not represent desirable redistricting strategy. Each of the major plans submitted to the Court also split several municipalities. While many of these divisions were made along county lines, we feel that consideration should have been given to the significance of the county lines within those communities. Dividing a congressional district along these lines is logical only for the sake of convenience. When the divided municipality is located in a densely populated area, we believe that the citizens identify more strongly with the services provided by the municipality. As a result, the principle of fair and effective representation would be better served in these circumstances by maintaining the integrity of municipal boundaries. After carefully examining all of the proposed plans under our second non-constitutional category, no plan can claim any particular advantage over the others. For every desirable element in a proposed plan, there is an equal number of undesirable features. We therefore turn to our third non-constitutional category for additional direction. “No one denies that a concept of ‘community of interest’ can and does apply to congressional redistricting, since formulating a plan without any such consideration would constitute a wholly arbitrary and capricious exercise.” Carstens Closing Argument, p. 3. Disputes about this third category center, instead, on the definition of the term “communities of interest” and its relevance to the State of Colorado. For our purposes, communities of interest represent distinctive units which share common concerns with respect to one or more identifiable features such as geography, demography, ethnicity, culture, socio-economic status or trade. We are convinced that a plan which provides fair and effective representation for the people of Colorado must identify and respect the most important communities of interest within the state. There is substantial agreement among the parties that Colorado should have a consolidated eastern agricultural district. Thus, the Governor’s Plan, which divides the state’s eastern counties into two separate districts is immediately placed at a considerable disadvantage. While the remaining plans essentially unite all of the eastern counties, there are a few additions and exclusions which detract from the concept of a consolidated agricultural district. H.B. 1624, for example, fails to include the two prominent southeastern agricultural counties of Crowley and Otero in its eastern plains district. The plan similarly ignores the agricultural concerns of eastern Arapahoe and Adams Counties. The Goens Plan, on the other hand, places Jackson County in its eastern plains district when this mountainous area has few, if any, common interests with rural farming communities. The parties also agree that the predominantly mountainous counties in the western portion of the state should be consolidated, if at all possible. Ideally, this western slope district would include only counties west of the Continental Divide. Unfortunately, population standards require that some front range counties be placed in this district. The choices made by the drafters of each plan submitted to the Court in this regard reflect policy decisions which often conflict with court-adopted criteria. The McPhee Plan # 2 was presented to the Court as a plan which truly respected geographic communities of interest within the state. See Exhibit 41. McPhee’s theory mirrors the ideal in that it attempts to group all mountainous areas west of the front range into one district. As a result, the McPhee Plan places a small portion of every major front range county into a western slope district. Testimony at trial demonstrated, however, that mountains are only important as they affect people. The people east of the Continental Divide have some very different concerns which frequently conflict with those of the people who reside on the western slope Moreover, in its attempt to respect geographic communities of interest, the McPhee Plan divides more counties than any other plan submitted to the Court and also achieves the largest population deviation. While we believe that communities of interest are an important factor in drawing fair and effective congressional districts, we are not prepared to recognize this criterion to the exclusion of all others. H.B. 1624 also reaches well into the front range to achieve the population necessary for a western slope district. The most objectionable feature of this plan is the manner in which it splits Pueblo. Carstens plaintiffs contend that Pueblo, or a portion of it, should be placed with El Paso County, its northern neighbor. We are not convinced, however, that this arrangement is in the best interests of the people of either county. These two areas share no strong communities of interest with the exception of the transportation concerns generated by a common highway and rail system. Pueblo is the only heavily industrial region in the state, while El Paso County supports a predominantly technical community. The current statistics relating to housing start-ups demonstrate that Pueblo is a low or negative growth county. El Paso, on the other hand, is a high growth area. Historically, these counties are commercial rivals and were characterized by the drafter of one plan as “hereditary enemies.” See Exhibit 41, p. 5. The competitive atmosphere between these two counties is contrary to the concept of communities of interest. We are persuaded that Pueblo and El Paso do not belong in the same congressional district. The Governor’s Plan leaves Pueblo intact and places it in a western slope district with the San Luis Valley. Testimony at trial indicated that Pueblo had considerably more amicable ties to the western slope than any other front range population center. The county’s large Hispanic population has strong traditional ethnic and cultural bonds with the San Luis Valley to the southwest. Pueblo is the primary trade, religious and educational center for the Valley which supplies much of Pueblo’s labor force. Because Pueblo is not experiencing rapid growth, its water requirements are stable, and conflict with the western slope over transmountain diversion of water would be kept to a minimum. The western slope district of the Governor’s Plan has at least one major disadvantage, however, in that it includes the front range counties of Gilpin and Clear Creek. While both of these counties are mountainous, their primary ties are with Boulder, Jefferson and Denver in the east. For example, both counties are included in a judicial district with their eastern neighbors. We are of the view that an acceptable redistricting plan should respect the front range orientation of Gilpin and Clear Creek Counties. After examining each of the plans with respect to our criteria, we have not been able to select one plan which substantially complies with all of our objectives. Each of the proposed plans appears to satisfy constitutional requirements. On balance, no one plan represents the best effort at providing fair and effective representation for the people of Colorado because each plan has several undesirable elements which tend to outweigh any advantages. Therefore, the Court has fashioned its own plan which attempts to satisfy the relevant legal criteria and incorporate the most desirable aspects of each plan presented to the Court. VI THE 1982 COLORADO CONGRESSIONAL REDISTRICTING PLAN We adopt the 1982 Colorado Congressional Redistricting Plan appended to this opinion and marked as “Exhibit A.” This plan creates the following districts: DISTRICT 1 incorporates the City and County of Denver with the exception of approximately 16,500 persons in the extreme southwest corner of the city. It also includes the Arapahoe County enclaves found within the Denver city limits. DISTRICT 2 includes Boulder, Gilpin and Clear Creek Counties, northeast Jefferson County, and western Adams County. The cities of Arvada, Broomfield, Federal Heights, Northglenn, Thornton and Westminster are all left intact. DISTRICT 3 is the largest district in the 1982 Plan. In addition to every county west of the Continental Divide, it contains Jackson, Pueblo and southern Fremont Counties along with those southern counties which comprise the San Luis Valley. It does not include Las Animas County. DISTRIC