OPINION AND ORDER

PETTINE, Senior District Judge.

This 42 U.S.C. § 1983 case involves the reapportionment of the Rhode Island Senate. Plaintiffs represent the class of all registered voters in Rhode Island. Defendants are the Secretary of State and the Board of Elections of Rhode Island. The Select Senate Redistricting Committee intervened on behalf of the defendants.

1. Procedural History

On April 9, 1982 the Rhode Island General Assembly enacted a reapportionment statute based on the 1980 census. This statute repealed the 1974 reapportionment scheme and established new state senatorial and representative districts. Subsequently, Justice Bulman of the Rhode Island Superi- or Court found that the 1982 reapportionment act violated the state and federal Constitutions. He enjoined the Secretary of State from conducting senatorial elections until a constitutional redistricting statute was enacted. Licht v. Quattrocchi, C.A. No. 82-1494 (R.I.Sup.Ct. June 3, 1982). Justice Bulman’s decision was affirmed without opinion by the Rhode Island Supreme Court. Licht v. Quattrocchi, 449 A.2d 887 (R.I., 1982).

On July 8, 1982 the Rhode Island Senate convened in a special session and passed a resolution creating a seven member Select Senate Redistricting Committee to develop a new senatorial reapportionment plan. The Rhode Island General Assembly enacted the plan recommended by the Redistricting Committee on July 20, 1982. The Governor, however, vetoed this plan because of his concern that it would not be possible to institute it and still hold the 1982 senatorial elections on schedule. 1 The General Assembly then passed 82-H-9101 which revived the 1974 senatorial lines for use in the 1982 elections, and provided that the new senatorial reapportionment plan developed by the Redistricting Committee would take effect beginning with the 1984 elections. 2

*86 On July 30,1982 Jonathan K. Farnum and James W. Hayes, Jr., individually and on behalf of all registered voters in Rhode Island, filed a complaint in this Court challenging the use of the 1974 senatorial lines in the 1982 elections under both the state and federal Constitutions. The parties agreed that, in light of the 1980 census, the 1974 district lines violated the one-person, one-vote principle of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964). They disagreed as to whether the impending senate election should and could proceed under the unconstitutional 1974 lines to avoid disruption of the state’s normal electoral processes. See e.g., Roman v. Sincock, 377 U.S. 695, 84 S.Ct. 1449, 12 L.Ed.2d 620 (1964); Cosner v. Dalton, 522 F.Supp. 350, 364 (E.D.Va.1981) (three-judge court). This Court acknowledged that equitable principles may require a court not to interfere with the conduct of rapidly upcoming elections where the election machinery is already in gear. 3 Here, however, it found that the election machinery was stalled and that the election would not be able to go forward as scheduled if the 1974 lines were used. It, therefore, enjoined the State Board of Elections from conducting the 1982 senatorial elections until a constitutionally permissible reapportionment scheme was devised. Farnum v. Burns, 548 F.Supp. 769, 775 (D.R.I.1982).

*87 Defendants have now filed a “Motion for Instructions and Other Relief.” They seek to determine whether the reapportionment plan developed by the Redistricting Committee and enacted by the General Assembly for use beginning in 1984 can be used in the special senatorial election made necessary as a result of the Legislature’s unsuccessful attempts to reapportion the State Senate. Plaintiffs oppose this motion on the grounds that defendants’ proposed interim plan violates the Equal Protection Clause of the Federal Constitution and the Nineteenth Amendment to the Rhode Island Constitution. Plaintiffs have submitted their own proposed interim reapportionment plan and request that it be used in the forthcoming special election.

In order for a plan to pass muster, it must meet two basic criteria; the one-person, one-vote requirement of federal law and the Rhode Island constitutional mandate that each senate district be as compact as possible and not the result of a political gerrymander.

II. Findings of Fact

Several facts are undisputed and are important in evaluating both the defendants’ and plaintiffs’ proposed interim plans. First, the 1980 census reveals that the population of Rhode Island is 947,154. Since the Rhode Island Constitution provides that the State of Rhode Island shall be divided into fifty senatorial districts, this means that the “ideal” senate district will contain 18,-943 persons. 4 Second, Providence’s population decreased by approximately 22,000 persons between 1970 and 1980, As a result of this population loss, Providence is no longer entitled under the one-person, one-vote principle of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), to the ten districts it was allotted under the 1974 reapportionment scheme. Rather, the Providence lines must be redrawn to comprise eight full districts and a portion of a ninth.

A. Defendants’ and Intervenors’ Plan

The plan proposed by defendants and intervenors for the special senatorial election is, as noted earlier, identical to that enacted by the General Assembly for use in elections in 1984 and thereafter. Dr. Henderson, a professor of political science and director of the computing center at Earlham College in Richmond, Indiana, stated that the total deviation under this plan was 5.61 percent and that the average deviation was 1.52 percent. 5 He further testified that it would have been possible to draw districts with smaller deviations and at the same time respect natural, historic, geographic and political boundaries.

The evidence presented at trial made clear that defendants, in formulating the plan now before us, used a “population window.” 6 Out of a total of fifty districts, thirty-five were developed as part of the initial reapportionment scheme considered by Justice Bulman in Licht v. Quattrocchi, C.A. No. 82-1494 (R.I.Sup.Ct. June 3, 1982), aff’d, 449 A.2d 887 (R.I., 1982). Justice Bulman found that the Legislature used a 2.5 percent population window in drawing these districts. At trial, this Court held that defendants were collaterally estopped from challenging this factual finding. 7 The *88 remaining 15 districts in the defendants’ plan comprise the City of Providence and its surrounding areas. These districts were developed after Justice Bulman struck down the initial senatorial reapportionment statute. We find that the defendants used a 5 percent population window in formulating the Providence districts. This finding is based on the computer printout summarizing the population statistics of the defendants’ plan and the testimony of Mr. Kimball Brace, the demographic expert, responsible for developing the defendants’ Providence districts.

Plaintiffs also presented evidence supporting their state law claim that the defendants’ Providence districts deviate from natural, historic, geographic and political boundaries for the purpose of achieving a political gerrymander. At trial, they demonstrated that the defendants’ plan does not respect the boundaries of the East Side of Providence. The East Side is a distinct geographic area which had comprised old districts 2 and 3 under the 1974 reapportionment scheme. See Ex. 1. Under defendants’ plan, however, old districts 2 and 3 were combined into a single district and the remainder of their population was dispersed into old districts 1, 4 and 8. This resulted in the homes of Senator Licht and Senator Sapinsley being located in the same district. See Ex. 2. Senators Licht and Sapinsley are the only Providence incumbents who are combined in one district under defendants’ plan. Indeed, Senator Sapinsley’s residence is located on the southernmost street of a small, rectangular section of district 2 (about five blocks long and two to four blocks wide) that protrudes into district 8.

Given the relatively stable population of the East Side, it is wholly illogical to combine old districts 2 and 3. The parties stipulated that these districts did not lose as much population as most of the other Providence districts. 8 District 2 needed only 574 additional people to equal an “ideal district,” of 18,943 persons, while district 3 was 2,550 below the norm. By contrast, districts 1 and 8 were 4,479 and 5,456 people below the norm. We credit Dr. Henderson’s testimony that the most logical way to account for the loss of population in Providence was to eliminate either districts 1 or 8 and disperse its population into surrounding districts.

Combining old districts 2 and 3 is also illogical because it ignores well recognized neighborhood lines. We credit the testimony of Professor Chester Smolski, the director, of the Geography and Urban Studies Program at Rhode Island College, that there are 25 neighborhoods in the City of Providence. Professor Smolski defined a neighborhood as “a specific territorial group. [It is] specific in the sense of physical identity of the area, and specific in the sense of social cohesiveness of the population in that area.” While some of these neighborhoods are in transition and the dividing line between them is sometimes arbitrary, each neighborhood does, in general, represent an independent set of interests or an independent geographic area.

The testimony at trial demonstrated that the defendants’ Providence districts were inconsistent with the neighborhood lines of the East Side, which have been respected since 1928, when the City of Providence was divided into senatorial districts for the first time. Although the East Side neighborhoods — Hope, Mount Hope, Blackstone, Wayland, College Hill and Fox Point — are linked geographically and surrounded by distinct boundaries, the defendants’ plan places portions of the East Side in four different districts. In doing this, the defendants’ plan divides the Hope and Mount *89 Hope neighborhoods and places sections of each in district 4, even though both neighborhoods are physically separated from that district by the so-called “Great Divide.” The “Great Divide” is a large, impassable barrier which consists of Interstate 95, the North Burial Grounds, a railroad yard and an industrial park. The East Side residents incorporated into district 4 include the residents of a large Senior Citizens’ development, many of whom moved into this complex from other addresses on the East Side.

The defendants’ plan also severs sections of the Fox Point and College Hill neighborhoods from the East Side. These neighborhoods have historically been considered part of the East Side and share many of the same interests as surrounding East Side neighborhoods. Under defendants’ plan they are linked with neighborhoods from which they are physically separated and with which they have little in common.

Plaintiffs also introduced testimony that district 8 of the defendants’ plan divides the Upper South Providence neighborhood and thereby dilutes the voting strength of the blacks who live in that neighborhood. It excludes the northern section of the Upper South Providence neighborhood, and instead includes a section of the South Elm-wood neighborhood which is largely comprised of white voters. Although defendants assert that their plan increases the minority population of adjacent district 9 from 43.19 percent to 46.98 percent, on our view of the evidence we need not resolve that factual issue. 9

We do find that defendants’ district 8 largely ignores historical, natural and geographical lines. Neither its northern nor southern borders follow logical geographic boundaries. It annexes the Fox Point neighborhood historically identified with the East Side. As a result, its population is concentrated in two sections that are separated by a large sparsely populated area, the downtown area and the Providence River.

Finally, the plaintiffs presented testimony demonstrating that the Redistricting Committee drew the Providence districts in the defendants’ plan for the purpose of accomplishing a political gerrymander. Both Senator Licht and Senator Sapinsley testified that the purpose of combining their districts was to preserve the senatorial districts of senators closely aligned with the democratic leadership at the expense of senators independent of that leadership. Senator Licht characterized Senator Sapinsley as “a very effective minority leader who is outspoken, who is quick to call to task the Democratic leadership” and who is “a whistleblower.” Senator Sapinsley testified that Senator Licht “is what I characterize as an independent Democrat. Others call him a maverick. He does not follow the leader blindly. He does not vote with the majority all the time as some members of the majority party do.” Both senators agreed that in contrast to themselves Senators Orabona and Hickey, who represent districts 1 and 8 in the defendants’ plan, are faithful to the democratic party leadership.

Senators Licht and Sapinsley both concluded that the defendants’ Providence districts were drawn for the purpose of accomplishing a political gerrymander. As Senator Sapinsley testified: “The gerrymander occurred because the districts were drawn without regard to natural, historic and political boundaries in order to reward or punish incumbents. In this case, the Democrat plan rewards the democrats who follow the leader, and punishes me, a republican.”

B. Plaintiffs’ Plan

To assist the Court in developing a court-ordered plan for use in a special election, plaintiffs submitted a senatorial reapportionment plan for the State of Rhode Island developed by Dr. Henderson. The total deviation under his plan is 1.58 percent and the average deviation is 0.26 percent. That plan eliminates district 8 of the 1974 reapportionment plan and disperses its popula *90 tion into surrounding districts. Professor Henderson adopted this approach because district 8 had suffered a substantial population loss between 1970 and 1980 and because its location in the center of Providence made it possible to disperse its population, while at the same time respect natural, geographic, historic and political boundaries.

We find that Henderson’s plan does in fact respect natural, geographic and historic boundaries. It preserves the identity of the East Side of Providence. It does not divide the Upper South Providence neighborhood and therefore does not dilute the voting strength of the blacks that live in that neighborhood. Moreover, the boundary between districts 2 and 8 in the Henderson Plan is a natural and geographic divide consisting of Route 195, the Providence River and an industrial area. Similarly, whenever possible, Route 95 serves as a district line. The plan includes no districts internally divided by large, sparsely populated areas.

At trial, the defendants attempted to show that the Henderson Plan was flawed and could not be implemented based on the documents submitted into evidence. Dr. Henderson admitted that in developing his plan, he had split certain voting districts and census blocks. He further testified that he had not allocated accurately the population in these split districts, and that this resulted in his incorrectly calculating the population variances of these districts. We find, however, that his population variances were only minimally inaccurate. 10 We further find that the Providence districts in the Henderson plan are sufficiently complete to be used in the forthcoming special senatorial election.

III. Conclusions of Law

A. Nineteenth Amendment to the Rhode Island Constitution

Plaintiffs’ state law claim is based on the Nineteenth Amendment to the Rhode Island Constitution. That amendment provides that senate districts shall be “as compact in territory as possible.” R.I. Const. Amendment 19, § 1. Plaintiffs claim that the Providence district lines in the defendants’ proposed interim reapportionment plan violate the principal of compactness. 11

The Rhode Island Supreme Court interpreted the compactness requirement in its Opinion to the Governor, 101 R.I. 203, 221 A.2d 799 (R.I.S.Ct.1966). It concluded that the compactness requirement was intended “to provide an electorate with effective representation" and prevent “the political gerrymander.” Id. at 802. The court defined the political gerrymander as follows:

*91 [It is] effected by a delineation of the boundaries of electoral districts in such a manner as to exclude therefrom, or to include therein, groups of voters whose political affinities may reasonably be surmised and whose political action is reasonably predictable. The delineation is usually without reference to physical features of the area that normally would induce regularity in configuration and bears no rational relationship to historical, natural and political boundary lines which normally would be relevant in the structuring of such districts.

Id.

The Opinion to the Governor suggests that a two-part test should be applied in determining whether the compactness requirement is violated. First, it must be determined whether district lines deviate from natural, historical, geographical and political lines. Second, assuming that such a deviation exists, it must be justified by rational and legitimate considerations, made in good faith, and free from any taint of arbitrariness or discrimination. Id. at 803.

In the present case, there can be no doubt that the defendants’ proposed Providence districts deviate from natural, historical, geographical and political lines. The geographical deviations are readily apparent from even a superficial glance at a map of Providence. Several district lines wholly ignore significant physical barriers. For example, the line separating district 2 and district 4 ignores the “Great Divide” and cuts off small sections of the Hope and Mount Hope communities from the East Side. Similarly, the College Hill and Fox Point neighborhoods are annexed onto old district 1 and district 8, even though they are physically separated from those districts by the state capítol grounds and the downtown area.

The defendants’ Providence districts not only violate the geographical boundaries of the East Side, but also violate the integrity of the East Side as an historical and political unit. As Professor Smolski testified, since 1928 the East Side has constituted a discrete political unit, the boundaries of which have always been respected. There is no merit to the defendants’ argument that the 1980 census made it necessary to combine sections of the East Side with other Providence districts. This suggestion cannot be reconciled with the Stipulation of the parties that between 1970 and 1980 the population of the East Side declined to a much lesser extent than that of other Providence districts.

The fact that the defendants’ proposed Providence districts deviate from geographical, historical and political lines does not in itself mean that they are constitutionally prohibited by the mandate of compactness. Additionally, it must be shown that the district lines were drawn for the purpose of achieving a political gerrymander. The Court is convinced that a political gerrymander did in fact occur in this case. There is no legitimate, nondiscriminatory explanation for the way in which the Select Senate Redistricting Committee redrew the Providence district lines. Placing sections of the East Side into four separate senatorial districts simply cannot be justified in view of that areas’ distinct boundaries and stable population. In fact, the only plausible explanation of the defendants’ proposed Providence district lines and the particular configuration of the southern boundary of district 2 is that they were drawn for a specific political purpose. As the testimony of Senators Licht and Sapinsley indicates, that purpose was to preserve the senatorial seats of incumbents favorable to the democratic leadership and punish incumbents critical of that leadership. 12

*92 This Court, therefore, holds that defendants’ proposed Providence district lines constitute a political gerrymander in violation of the Nineteenth Amendment to the Rhode Island Constitution. In reaching this decision, the Court is mindful of the fact that redistricting is a legislative activity in which courts are generally reluctant to intervene. This case, however, presents the Court with no alternative. The Constitution of the State of Rhode Island provides that only by prohibiting the political gerrymander is it possible to ensure effective participation in representative government. Opinion to the Governor, 221 A.2d at 802. The Redistricting Committee’s attempt to reapportion the City of Providence for its own political purposes threatens that guarantee. Accordingly, the Court cannot sanction the use of the Providence district lines it proposes in the forthcoming special election.

B. Federal Constitutional Issues

This Court’s conclusion that the defendants’ proposed Providence districts violate the Nineteenth Amendment to the Rhode Island Constitution does not require it to invalidate the defendants’ entire redistricting plan. Reapportionment is primarily a matter for the state legislature, see Gaffney v. Cummings, 412 U.S, 735, 750-51, 93 S.Ct. 2321, 2330, 37 L.Ed.2d 298 (1973); White v. Weiser, 412 U.S. 783, 795-96, 93 S.Ct. 2348, 2354-55, 37 L.Ed.2d 335 (1973), and this Court must not intrude upon state policy any more than necessary. Whitcomb v. Chavis, 403 U.S. 124, 160, 91 S.Ct. 1858, 1877, 29 L.Ed.2d 363 (1971). Where, as here, only a portion of the plan proposed by a state legislative committee is unconstitutional under state law, the Court believes that it would be inappropriate to strike down the entire plan. 13 It is thus necessary to consider whether the thirty-five districts outside the City of Providence are valid under the Federal Constitution.

At trial, this Court held that the defendants were collaterally estopped from challenging Justice Bulman’s finding that a 2.5 percent population window was used in formulating the thirty-five districts outside the City of Providence. Plaintiffs argue that the defendants’ use of a population window in reapportioning the State Senate violated the Equal Protection Clause of the United States Constitution. To support this contention, they rely primarily on Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519 (1969), in which the Supreme Court stated:

We reject Missouri’s argument that there is a fixed numerical or percentage population variance small enough to be considered de minimis and to satisfy without question the “as nearly as practicable” standard. The whole thrust of the “as nearly as practicable” approach is inconsistent with adoption of fixed numerical standards which excuse population vari *93 anees without regard to the circumstances of each particular case. The extent to which equality may practicably be achieved may differ from State to State and from district to district. Since “equal representation for equal numbers of people [is] the fundamental goal for the House of Representatives,” ... the “as nearly as practicable” standard requires that the State make a good-faith effort to achieve precise mathematical equality.

Id. at 531, 89 S.Ct. at 1229.

This Court does not agree that the defendants’ use of a population window violated the Equal Protection Clause. The Supreme Court has consistently recognized that reapportionment plans for the state legislature are not subject to the same strict standards applicable to congressional reapportionment plans. See e.g., White v. Regester, 412 U.S. 755, 763, 93 S.Ct. 2332, 2338, 37 L.Ed.2d 314 (1973); Mahan v. Howell, 410 U.S. 315, 321-26, 93 S.Ct. 979, 983-86, 35 L.Ed.2d 320 (1973). Furthermore, it is now well established that population variances of less than ten percent in state reapportionment plans are of prima facie constitutional validity. See Connor v. Finch, 431 U.S. 407, 418, 97 S.Ct. 1828, 1835, 52 L.Ed.2d 465 (1977); White v. Regester, 412 U.S. at 764, 93 S.Ct. at 2338. In light of the deference that must be afforded to state reapportionment plans, we hold that when such a plan has a total deviation of less than ten percent, it does not violate the one-person, one-vote principle merely because a population window was used in developing that plan.

Kirkpatrick does not, as the plaintiffs suggest, require that this Court reach a different result. Kirkpatrick held that the use of a population window in drawing congressional districts was inconsistent with the obligation to attempt to achieve absolute population equality. In developing state legislative districts, however, the state is not obligated to attempt to achieve absolute population equality. Rather, it may seek to ensure “some voice to political subdivisions, as political subdivisions,” while at the same time ensuring that there exists “substantial equality of population among districts.” See Mahan v. Howell, 410 U.S. at 321-22, 93 S.Ct. at 983-84; Reynolds v. Sims, 377 U.S. at 579-81, 84 S.Ct. at 1390-91. Because the state may use a population window and still comply with its obligation to obtain “substantial” population equality, the Court declines the plaintiffs’ invitation to extend Kirkpatrick to invalidate the use of population windows in reapportioning state legislatures. Therefor, we conclude that the thirty-five districts of the defendants’ plan which are outside the city of Providence and its surrounding areas pass constitutional muster.

IV. Remedy

Having concluded that the defendants’ proposed Providence districts violate the Nineteenth Amendment of the Rhode Island Constitution, the Court must now adopt a court-ordered plan for these districts. The Supreme Court has held that a court-ordered reapportionment plan must achieve the goal of population equality with only a de minimis variation. See Connor v. Finch, 431 U.S. 407, 414, 97 S.Ct. 1828, 1833, 52 L.Ed.2d 465 (1977); Chapman v. Meier, 420 U.S. 1, 26, 95 S.Ct. 751, 765, 42 L.Ed.2d 766 (1975). It has further held that a court must justify any deviation from approximate population equality on the basis of historically significant state policy or unique features. See Chapman v. Meier, 420 U.S. at 26, 95 S.Ct. at 765.

This Court finds that the Providence districts of the Henderson Plan satisfy the standards used to evaluate the constitutionality of court-ordered redistricting plans. The total deviation of the Henderson Plan is 1.58 percent and the average deviation is 0.26 percent. These deviations are de minimis, and, in any event, are attributable to Dr. Henderson’s attempt to respect the natural, geographic and historic *94 boundaries of the City of Providence. The Court is not convinced, as the defendants contend, that Henderson’s plan is inaccurate and unworkable. The deficiencies in the Henderson Plan are minor and do not call into question the overall accuracy of the plan as it was presented to this Court. Accordingly, we find the Providence districts of the Henderson Plan to be the only constitutional districts presented to us for that area and so adopt these districts for use in a special senate election.

The Court holds that the State Board of Elections shall conduct the special senatorial elections under a reapportionment plan that includes both the Providence districts of the Henderson Plan and the thirty-five districts outside the City of Providence in the defendants’ plan. We recognize that there may be some difficulty in merging the two plans. Their merger will require keying the 6 districts surrounding the City of Providence to both Henderson’s Providence districts and the thirty-five districts of the defendants’ plan. We, therefore, order the parties to meet and attempt to develop such a reapportionment scheme and submit it to the Court within five days. If such a scheme cannot be developed, counsel will immediately notify this Court. In such event, the Court will devise a reapportionment plan in accordance with the principles spelled out in this Opinion.

*95 [[Image here]]

*96 [[Image here]]

. Prior to vetoing the senatorial reapportionment plan recommended by the Redistricting Committee, the Governor petitioned the Rhode Island Supreme Court for an advisory opinion as to whether Justice Bulman’s decision declaring the first reapportionment act unconstitutional revived the 1974 district lines. A majority of the Justices of the Supreme Court concluded that “until the General Assembly acts to redistrict the Senate in accordance with the terms of the judgment of the Superior Court, in order to avoid a vacuum in the electoral process relating to the Senate, it is our determination that the [1974 district lines] are and continue to be valid and subsisting.” See Licht v. Quattrocchi, 454 A.2d 1210, 1211 (R.I., 1982).

. Prior to passage of this Act, the Rhode Island Supreme Court, upon motion of the Attorney General, lifted the injunction entered by Justice Bulman in order to permit the State Board of Elections to conduct a senate election in accordance with the 1974 district lines. In entering this order, the Supreme Court relied upon: “(1) the assertions of counsel that the 1974 districts were constitutionally unchallenged and therefore valid, (2) the agreement of all parties to the litigation that it was constitutionally permissible and physically feasible to hold the 1982 election for the state senate under district lines established in 1974, and (3) the apparent agreement of counsel that the public interest would best be served by going forward with an election under the 1974 senate dis *86 tricts.” Licht v. Quattrocchi, 454 A.2d at 1212 (R.I., 1982). These representations were made to the Rhode Island Supreme Court at a hearing on the Governor’s request for an advisory opinion as to whether Justice Bulman’s decision declaring the first reapportionment statute unconstitutional revived the 1974 district lines.

. We cannot accept the Rhode Island Supreme Court’s conclusion that this Court should not have assumed jurisdiction in this case. Licht v. Quattrocchi, 454 A.2d 1210 (R.I., 1982). First, even if we had known the basis of the Rhode Island Supreme Court’s decision to lift Justice Bulman’s injunction, we do not believe the plaintiffs were barred by the doctrine of res judicata from challenging the constitutionality of the 1974 district lines. Res judicata only applies to issues that could have been fully and fairly litigated in a prior proceeding. See Allen v. McCurry, 449 U.S. 90, 101 S.Ct. 411, 66 L.Ed.2d 308 (1980). Here, plaintiffs were not provided with a prior opportunity to fully and fairly litigate the constitutionality of using the 1974 lines in the 1982 senatorial election. The fact that plaintiffs’ counsel made certain representations to the Rhode Island Supreme Court concerning this issue does not suggest otherwise. The representations of counsel were made at a hearing concerning the Governor’s request for an advisory opinion on the narrow question of whether Justice Bulman’s decision striking down the first senatorial reapportionment statute revived the 1974 district lines. Specifically, the representations were made in response to a request by the Rhode Island Supreme Court for counsels’ “reaction” to a “proposal” by the Attorney General to lift Justice Bulman’s injunction in order to let the 1982 elections go forward under the 1974 lines. Licht v. Quattrocchi, 454 A.2d at 1212 (R.I., 1982). Counsel were not given a full and fair opportunity to litigate an issue when they were asked for their “reaction” to a “proposal.”

Second, abstention would have been inappropriate in this case. The plaintiffs’ challenge was to the constitutionality of using the 1974 district lines in the 1982'state senatorial elections. Their complaint did not raise an unsettled issue of state law. See Railroad Commission of Texas v. Pullman Co., 312 U.S. 496, 61 S.Ct. 643, 85 L.Ed. 971 (1941). Nor was there a pending state court suit involving reapportionment of the state senate at the time plaintiffs filed their complaint in this Court. See Colorado River Water Conservation District v. United States, 424 U.S. 800, 96 S.Ct. 1236, 47 L.Ed.2d 483 (1976).

This simply was not a case like Scott v. Germano, 381 U.S. 407, 85 S.Ct. 1525, 14 L.Ed.2d 477 (1965), in which the state judiciary was in the process of requiring valid reapportionment. At the time the plaintiffs filed their complaint, the Rhode Island Supreme Court had lifted the injunction entered by Justice Bulman and thereby relinquished jurisdiction over the controversy. It appeared that the controversy was settled and that the elections would proceed under the 1974 district lines. A lawsuit was subsequently filed in this Court, however, challenging the use of the 1974 lines. There can be no doubt that this challenge presented a substantial federal question; the parties agreed that the use of the 1974 lines in the 1982 senate elections violated the one-person, one-vote requirement of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964). In addition, this Court was faced “on the one hand with the definite problem that if the clearly unconstitutional senate lines are implemented, there will be no election at all as scheduled, and on the other hand with the certainty that, if the senate elections are enjoined, the others will proceed on time.” Farnum v. Burns, 548 F.Supp. 769, 775 (D.R.I.1982). Under such circumstances, it would have been inappropriate for this Court to abstain.

. The Supreme Court has assessed the constitutional validity of reapportionment schemes under the one-person, one-vote principle by comparing the population of the actual districts under challenge to that of a hypothetical “ideal district.” The population of an “ideal district” is determined by dividing the total population of a state by the number of senatorial districts.

. The total deviation of a reapportionment plan is determined by adding the deviation of the district with the largest population to the deviation of the district with the smallest population. Average deviation is calculated by averaging the deviation of all of the districts.

. The use of a “population window” ensures that the population of all districts in a redistricting plan will fall within a predetermined range, above and below, the ideal district.

. This Court, however, is not bound by Justice Bulman’s legal conclusion that use of a 2.5 percent population window violates the obligation to develop state legislative districts that are as nearly equal in population as possible. The Rhode Island Supreme Court did not spe *88 cifically affirm Justice Bulman on this point. See Licht v. Quattrocchi, 449 A.2d 887 (R.I., 1982). Accordingly, we are only bound by Justice Bulman’s findings of fact on this issue.

. The defendants contend that no evidence was introduced at trial as to the changes in population in the Providence districts from 1970 to 1980. Post-Trial Brief of Defendant-Intervenors at 57. This claim is without merit. Counsel for the intervenors admitted at trial that there was no dispute as to the population changes and agreed to stipulate to these facts. (Tr. Vol. I at 45-47).

. Plaintiffs contest the accuracy of the defendants’ minority percentage figures on the ground that they include minorities other than blacks and hispanics. Because we do not reach the plaintiffs’ minority vote dilution claim, we also need not resolve this factual question.

. Defendants demonstrated that Dr. Henderson split a number of census blocks — the largest of which allegedly contained 205 people. Dr. Henderson testified, however, that even assuming that he split a census block containing 205 people without accurately allocating its population, the population variance which he calculated for the affected districts would only change by one percent, and the average deviation for his entire plan would only change by one fiftieth of one percent.

. Defendants contend that their proposed interim redistricting plan constitutes a “presently existing statutory redistricting plan,” and that, consequently, it should be afforded a presumption of constitutionality. Post-Trial Brief of Defendant-Intervenors at 6-14. This Court disagrees with the defendants’ contention that the district lines enacted by the Legislature for use in 1984 are presently effective. The severability clause of the reapportionment statute governing the 1982 and 1984 elections does not suggest otherwise. The severability clause provides only that if any provision of the statute is held unconstitutional, “the remaining provisions ... shall not be deemed to be impaired or effected thereby but shall be deemed to remain in full force and effect.” R.I.G.L. 82-H-9101(14). Neither the severability clause nor any other part of the statute provides that the 1984 lines will become immediately effective should the 1974 lines be found unconstitutional. Defendants’ argument that the severability clause should be interpreted to reach this result is to no avail. The severability clause only requires this Court to avoid striking down the entire reapportionment statute because one of its provisions is found unconstitutional. In the present case, this means that the 1984 lines remain in effect for 1984; it does not mean that the 1984 lines should be used in 1982. To hold otherwise, would be to interpret the severability clause in a way that is not supported by its plain language.

. Defendants argue that there was no political gerrymander in this case even if the Redistricting Committee drew the Providence district lines to ensure the reelection of certain incumbents. They claim that “it is voters ..., and not incumbents or candidates, who must be found to be victims if there is to be a political gerrymander. Stated differently, the political gerrymander is simply another facet of vote dilution.” Post-Trial Brief of Defendant-Intervenors at 51. The defendants’ argument must be rejected. Here, it makes no difference whether voters or incumbents are viewed as the “victims” of a political gerrymander since *92 both are equally affected by the defendants’ proposed Providence districts. The. Court is convinced that the defendants’ Providence districts were drawn to exclude East Side voters from the districts of Senators Licht and Sapinsley to ensure the reelection of incumbents faithful to the democratic leadership. This is precisely the type of political gerrymander prohibited, by the Nineteenth Amendment to the Rhode Island Constitution. See Opinion to the Governor, 101 R.I. 203, 221 A.2d 799, 802 (R.I. S.Ct.1966).

. Deference to the defendants’ plan is particularly appropriate in light of this Court’s ruling at trial that it is a legislative plan. This conclusion was based on the United States Supreme Court’s decision in McDaniel v. Sanchez, 452 U.S. 130, 101 S.Ct. 2224, 68 L.Ed.2d 724 (1981). McDaniel held that a redistricting plan that had been prepared by a legislative consultant under the direction of local officials, but had not been enacted by the municipal body, was a legislative plan for purposes of the preclearance requirements of section five of the Voting Rights Act. Id at 153, 101 S.Ct. at 2238. The Court based this conclusion on the fact that the plan at issue was drafted under the supervision of municipal officials, and therefore reflected the policy choices of the elected representatives of the people. Here, as in McDaniel, the plan at issue was drafted by a legislative consultant under the direction of a legislative body. The defendants’ plan reflects the policy choices of the elected representatives and therefore must be considered as a legislative plan.