Baker v. Carr, 222 F. Supp. 684 (M.D. Tenn. 1963)

*685 Z. T. Osborn, Jr., Nashville, Tenn., Hobart F. Atkins, Knoxville, Tenn., Walter Chandler, Memphis, Tenn., for plaintiffs.

Neill S. Brown, Director of Law, Seymour Samuels, Asst. Director of Law, Metropolitan Government, Nashville, Tenn., Harris Gilbert, Nashville, Tenn., for City of Nashville (Metropolitan Government of Nashville and Davidson County, Tenn.).

C. R. McClain, Director of Law, Knoxville, Tenn., for City of Knoxville, Tenn.

J. W. Anderson, City Atty., Chattanooga, Tenn., E. K. Meacham, Eugene N. Collins, Chattanooga, Tenn., for City of Chattanooga, Tenn.

Cecil Branstetter, Nashville, Tenn., for Tenn. State Labor Council.

Gilbert S. Merritt, Jr., Asst. Director of Law, Metropolitan Government, Nashville, Tenn., for Davidson County Election Commission.

George H. McCanless, Atty. Gen., and Milton P. Rice, Asst. Atty. Gen., State of Tennessee, Nashville, Tenn., for defendants.

Edwin F. Hunt, Nashville, Tenn., for Tennessee Farm Bureau Federation.

Before WEICK, Circuit Judge, and BOYD and MILLER, District Judges.

BY THE COURT.

This action again comes before us for consideration of the problem of legislative apportionment in Tennessee.

On June 22, 1962, the Court rendered a per curiam opinion, Baker, et al., v. Carr, et al., 206 F. Supp. 341, setting forth its views with respect to the 1962 legislation enacted by an extraordinary session of the General Assembly of Tennessee when measured in terms of the equal protection requirements of the Fourteenth Amendment. It was the conclusion of the Court that although the Act apportioning seats in the House of Representatives possessed certain individual inequities which cast doubt upon its validity, the general plan followed by the Act was not per se irrational or arbitrary. The scheme of the Act was to apportion the 99 members of the House of Representatives on the basis of a modified voter population principle by applying the two-thirds provision of the Tennessee Constitution not only to counties but also to floterial districts. This method of distribution obviously resulted in favoring the less populous areas of the state vis-a-vis the metropolitan areas having concentrations of population. However, as set forth in the opinion, "[s]uch a state plan for distribution of legislative strength, at least in one house of a bicameral legislature, cannot, in our opinion, be characterized as per se irrational or arbitrary." 206 F. Supp. at 345. The Court went on to point out that this was true for the reason that the Fourteenth Amendment should not be construed to preclude a state from enforcing a policy which would give a measure of protection to its rural and sparsely populated counties constituting integral and historic parts of the state's governmental structure and possessing substantial interests in state government and in the formulation of its laws and policies.

Consideration of the 1962 Act apportioning seats in the state senate, however, led the Court to the conclusion that the Act was devoid of any standard or rational plan of classification which the Court was able to discern. 206 F. Supp. at 346. Discrepancies were found not only as between the three grand divisions of the state but as between rural areas and metropolitan areas, as between various rural districts, and even as between metropolitan districts. For the reason that the Act was wholly wanting in any rational or explainable principle or theory of representation, comparable to the "crazy quilt" referred to by Mr. Justice Clark in his concurring opinion in this case, 369 U.S. at 254, 82 S. Ct. at 729, 7 L. Ed. 2d 663, the Court expressed *686 the view that the Act did not meet the test of equal protection of the law.

Confronted with the problem of fashioning a remedy, the Court carefully weighed and considered various alternatives, 206 F. Supp. 348-349, and concluded that, rather than for the Court to undertake the delicate task of devising and effectuating a plan, the best solution was to permit the General Assembly itself to reconsider the problem with the help of certain guide lines which the Court undertook to prescribe in its opinion. These were set forth as follows:

"* * * we are of the opinion that if the two-thirds principle should be applied and used in apportioning seats in the House of Representatives, either by applying it to counties or floterial districts, or to both, then it would follow that the Senate would have to be apportioned on the basis alone of numbers of qualified voters. Contrariwise, if the Senate should be apportioned on an equitable and rational basis not fully reated to voting strength, it would be necessary to apportion the House of Representatives on the basis of numbers of qualified voters alone, without regard to the two-thirds principle whether applied to counties or districts or both.

"It is noteworthy that the 1962 statutes fail to apportion either house on the basis of qualified voters. In addition to individual inequities and inequalities present in both statutes, the net result is to perpetuate a `gross disproportion of representation to voting population.' We find in the context of this case that equal protection requires that such condition be eliminated and that apportionment in at least one house shall be based, fully and in good faith, on numbers of qualified voters without regard to any other factor." 206 F. Supp. 349.

Thus, the Court clearly indicated its view that the minimum standard required to satisfy equal protection in legislative apportionment in Tennessee could be achieved by a reasonable departure from the principle of per capita representation in one house of the General Assembly in order to protect the interests of the smaller or rural areas of the state, coupled with full recognition of such principle, insofar as practicable, in the other house. The Supreme Court in its opinion did not see fit or perhaps did not have occasion to prescribe any exact criteria for determining the meaning of equal protection of the law under the Fourteenth Amendment in the context of state legislative representation. But it was our view then that equal protection should not be construed in such way as to deny to a state the right in distributing its legislative strength to give a rational recognition to both geography and population. Such a view, we believed, found support in the legislative practice of many of the states as well as in the analogy of the United States Congress.

At its regular 1963 session the General Assembly of Tennessee again enacted apportionment legislation. Chapter 295 of the Public Acts of 1963 reapportions seats in the House of Representatives. Chapter 320 of the Public Acts of 1963 reapportions seats in the state senate. The plaintiffs, on September 6, 1963, filed an amended and supplemental complaint in which they challenge the constitutionality of both Acts upon the ground that they do not meet the minimum standards prescribed by this Court in its opinion of June 22, 1962, and upon the ground that both Acts are violative of the equal protection clause of the Federal Constitution. They request that both Acts be declared unconstitutional and that the Court approve plans suggested by them which they insist would apportion both houses of the General Assembly as nearly as possible according to the principle of per capita equality of representation. Their present insistence is that any substantial departure from numbers is forbidden in apportioning either house and that the Court should give effect to the so-called principle that "one person equals one vote." The defendants, on the other *687 hand, contend that the 1963 Acts are constitutionally unobjectionable and that they fully comply with the minimum standards indicated by this Court in its prior opinion. They point out that the 1963 Act reapportioning the House of Representatives eliminates all of the inequities specified by the Court but otherwise faithfully follows the general plan of the 1962 Act. They insist that the Act reapportioning the Senate is primarily and substantially based upon numbers of qualified voters as shown by the 1960 federal census.

First we consider the plaintiffs' insistence that we should retreat from the view expressed a year ago and enunciate a rule that the Federal Constitution rigidly requires the states in constituting their legislative bodies to be governed exclusively by population except for minor deviations due to the practical inconvenience of undertaking to disregard county boundary lines.

Upon reconsideration we adhere to our former view that while population must be substantially adhered to in one house of a bicameral legislature, it is permissible under the Fourteenth Amendment for a state in creating legislative districts for the other house to take into account other interests and objectives, as for example, the policy of providing adequate legislative representation to the rural population whose special interests might otherwise be overridden by "sheer weight of numbers." This view has been generally approved by other courts having occasion to consider the question. See Sobel v. Adams, 208 F. Supp. 316 (1962); Sims v. Frink, D.C., 205 F. Supp. 245 and D.C., 208 F. Supp. 431 (1962); Tombs v. Fortson, D.C., 205 F. Supp. 248 (1962); Lisco v. Love, D.C., 219 F. Supp. 922; Germano, et al., v. Kerner, et al., D.C., 220 F. Supp. 230 (1963); W. M. A. C. v. Simon, D.C., 202 F. Supp. 741 and D.C., 208 F. Supp. 368; Nolan v. Rhodes, D.C., 218 F. Supp. 953 (1963); Caesar v. Williams, 84 Idaho 254, 371 P.2d 241 (1962); Jackman v. Bodine, 78 N.J.Super. 414, 188 A.2d 642 (1963); Maryland Committee v. Tawes, 229 Md. 406, 184 A.2d 715 (1962); Sweeney v. Notte, 183 A.2d 296 (Rhode Island 1962).

It is true that the Supreme Court in Gray v. Sanders, 372 U.S. 368, 381, 83 S. Ct. 801, 9 L. Ed. 2d 821, stated that "[t]he conception of political equality * * * can mean only one thing one person one vote." It was made clear in the opinion of the Court, however, that the case involved state-wide elections of officers responsible to a state-wide constituency, and that the question had nothing to do "with the composition of the state or federal legislatures." It was expressly stated by the Court that "we intimate no opinion on the constitutional phases of that problem beyond what we said in Baker v. Carr, supra. The present case is only a voting case." That equal protection does not require that apportionment of both houses of a bicameral legislature be fully related to population appears also to have been supported in the concurring opinions of Justices Douglas, Clark and Stewart in the present case. Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691, 7 L. Ed. 2d 663. It may be that the issue will be definitely settled by the Supreme Court in cases now pending before it, but until a different rule is enunciated by our highest tribunal, we are content to hold that the Fourteenth Amendment does not require a state in legislative districting to recognize exclusively the interests of popular representation. Indeed, support for this view is found in the language of the Supreme Court in MacDougall v. Green, 335 U.S. 281, 69 S. Ct. 1, 93 L.Ed. 3:

"* * * To assume that political power is a function exclusively of numbers is to disregard the practicalities of government. Thus, the Constitution protects the interests of the smaller against the greater by giving in the Senate entirely unequal representation to populations. It would be strange indeed, and doctrinaire, for this Court, applying such broad constitutional concepts as due process and equal protection of the laws, to deny a State the power *688 to assure a proper diffusion of political initiative as between its thinly populated counties and those having concentrated masses, in view of the fact that the latter have practical opportunties for exerting their political weight at the polls not available to the former. The Constitution a practical instrument of government makes no such demands on the States. * * *"

An examination of Chapter 295 discloses that the individual inequities in the apportionment in the House of Representatives which gave the Court concern in 1962 have now been eliminated, leaving the apportionment of the House otherwise as provided by the 1962 legislation. It follows that if Chapter 320 of the Public Acts of 1963 constitutes an apportionment of seats in the Senate solely on the basis of numbers of qualified voters, the Court would have no reason to withhold its approval of the 1963 legislation in its entirety. Such an apportionment would be in compliance with the Court's view of the minimum requirements of the equal protection clause.

The next question therefore to be considered is: Does the 1963 Act apportioning seats in the Senate of Tennessee distribute such seats as nearly as practically possible on the basis of qualified voters alone? Consideration of this issue in the light of the 1963 legislation itself and the undisputed facts of the case compel us to answer this question in the negative. Admittedly the 1963 Act is an improvement over the Senate Act of 1962. It eliminates many of the "crazy quilt" features of the former act but it contains substantial deviations from the principle of voter population resulting in invidious discriminations against voters in at least three of the four metropolitan counties of the state.^[1]

On the basis of the 1960 Federal Census, the Senate ratio is 63,420 qualified voters. In awarding Senate seats to Shelby, Davidson, Knox and Hamilton Counties, one seat was given for each full ratio but fractions of a ratio in these counties were disregarded. Thus, Shelby County having a voting population of 359,532 was awarded five senators with no recognition being given to its fraction of 42,432 qualified voters. Davidson County with a voting population of 242,933 was awarded three senators but its very substantial fraction of 52,673 qualified voters was ignored a fraction which is only 10,747 voters short of a full additional ratio. Hamilton County was given two senators for its voting population of 142,979 qualified voters and a fraction of 16,139 was disregarded. Knox County with a voting population of 151,999 was given two senators and its fraction of 25,159 was also disregarded. Fractions in these four metropolitan counties represent approximately 136,000 qualified voters of the state. The reason that these fractions were treated in this manner is stated in the report of the Senate Committee as follows:

"The Constitution of Tennessee makes voting population, which is generally understood to mean citizens who are 21 years old or older, the basis for apportionment of State Senators. By such Constitution the membership of the Senate is limited to 33, and by statute there are now, and for many years have been, 33 Senate seats. The voting population of Tennessee is 2,092,891. If these voters lived in 33 districts of equal population, there would be 63,420 such voters in each district and such figure (63,420) is called `the ratio.' Because the Tennessee Constitution refers to voting population (`qualified electors') this Report does the *689 same. Actually, voting population is approximately 60 per cent of total population, and for every county except a very few, its percentage of total population is substantially the same as its percentage of voting population in the State.

"In Tennessee there are five counties, frequently called the metropolitan counties, which have a population justifying the apportionment to each of them of one or more senators to be elected exclusively by the voters of such county and to represent no other county. Your Committee decided that before creating senatorial districts composed of two or more counties, it was proper, if not necessary, to determine the number of senators to be apportioned to each of these most populous counties, being Shelby, Davidson, Knox, Hamilton and Sullivan Counties.

"Shelby County has a population in excess of five ratios, but less than six. Davidson County has a population in excess of three ratios, but less than four. Knox and Hamilton Counties each have a population in excess of two ratios, but less than three. Sullivan County has a population very slightly in excess of one ratio.

"There was no question, insofar as the Committee was concerned, that each of these metropolitan counties should receive the entire number of senators for which it had full ratios, but the problem was what, if anything, should be done about fractions of a ratio in each of these counties. Before stating the decision and recommendation of your committee in this regard, we should call attention to a constitutional provision and to certain practical considerations in respect to the matter.

"The Tennessee Constitution forbids that any county shall be divided in forming a senatorial district. As a practical matter this means that every voter in Shelby County will be entitled to vote for and participate in the election of five senators; and the voter in Davidson County in the election of three senators. This provision assures to the voters in such counties the practical opportunity to exert greater political weight by the election of a slate or ticket backed by a political organization and both supported and publicized by a metropolitan press. The voter in a multicounty district has no such opportunity.

"The provision of the Tennessee Constitution creating the county as an unbreakable unit in the selection of senators has another practical effect: It compels the result that the metropolitan county must be over-represented or under-represented to the extent of its fraction of a ratio. If a metropolitan county be given a full senator by reason of its fraction, there is clear over-representation. If such county, as compensation for its fraction, be placed in a senatorial district, the over-representation is just as real and practical, although theoretically there is no such favored treatment.

"For example, if Shelby County (1960 voting population: 359,532) is given a `share' in a sixth senator with Tipton County (1960 voting population: 14,912) the practical result has been demonstrated in the political history of Tennessee. Shelby County almost invariably would take, and in the past has taken, such `shared' senator. Under the Apportionment Act of 1901, Shelby County and Tipton County did `share' a senator. From 1920 through 1961, there was one General Assembly (1925) when such senator came from Tipton County. In every other instance the `shared' senator was from Shelby County. Of course, the reference to Shelby County is illustrative only and your Committee believes the same result would be no less a certainty for any other county.

*690 "Recognizing this fact, a number of the spokesmen for these metropolitan counties have expressed understanding of, and acquiescence in, the viewpoint that fractions of a ratio should be disregarded.

"Your Committee, therefore, decided that no small or medium-sized county should be attached to a metropolitan county in the creation of a senatorial district. Your Committee further decided that for counties having one or more full senators fractions of a ratio would not be considered. Thus it was determined that Shelby County should be apportioned five senators; Davidson County three senators; Knox and Hamilton Counties two senators each; and Sullivan County one senator. The result is that the five most populous counties, with an area of 2909 square miles, will be given thirteen senators; and there remain for ninety counties with an area of 39,335 square miles, a total of twenty senators."

It is thus abundantly clear that large fractions in metropolitan counties were disregarded simply because they were located in such metropolitan counties and that the Committee decided to take factors other than population into account. This approach naturally led to a substantial departure from the principle of per capita representation, injecting into the Senate plan of 1963 the same concept of area or small county representation which prevailed in apportioning the House of Representatives. The treatment given to the metropolitan fractions, as shown by the Committee Report itself, manifestly deviated from the direction of the Court that one house should be apportioned on the basis of qualified voters without regard to any other factor.

The amended complaint, in our opinion, correctly describes the 1963 Senate Act in the following terms:

"Plaintiff's Exhibit `D', showing the apportionment of the Senate of Tennessee, reflects the same plan and pattern of discrimination against the plaintiffs, and others similarly situated. The average voting population of all districts is 63,421. Senatorial district #21 has a voting population of 50,943 a ratio of .80 and is permitted to elect a senator; senatorial district #23 has a voting population of 50,645 ratio .80 and is permitted to elect a senator and senatorial district #5 has a voting population of but 50,887 ratio .80 and is allowed to elect a senator. Voting population per senator from Davidson County is 80,978, a ratio of voting population to average of 1.28; voting population of Shelby County per senator is 71,906, ratio 1.13; voting population of Knox County is 76,000 per senator, ratio 1.20, while the voting population of Hamilton County per senator is 71,490, a ratio of 1.13. Plaintiffs would show that ten of the thirty-three districts in the Senate have a ratio of .84 or less (53,506 voters or less) and that in the apportionment of districts in the Senate, the very same counties which have achieved the most advantageously discriminatory representation in the House are again favored in the Senate apportionment by being placed usually in those districts having a voting population ratio of .84, 53,506 voters, or less. The same four largest counties Davidson, Shelby, Knox and Hamilton, so seriously and purposely discriminated against in the House are again the counties most discriminated against in the Senate. These counties are the only counties placed in districts with a ratio of 1.13 or more, 71,940 voters or more; all other ninety-one counties are better represented in the Senate."^[2]

*691 It is recognized that exact equality cannot be achieved as long as the practice of allocating representation to counties and to districts composed of counties without the division of any county into districts is followed. Necessarily in making a population apportionment on this basis there will be fractions of a ratio in particular districts representing either an under-representation or an over-representation. But if the nearest possible per capita distribution is to be achieved substantial fractions of a ratio cannot be arbitrarily disregarded simply because they are located in metropolitan counties.

The problem is over-simplified by taking the view, as the Senate Committee evidently did, that the only question was what, if anything, should be done about fractions of a ratio in these counties. The real problem before the 1963 Legislature, if the House plan was to be approved, was to determine how the ideal of an equal per capita distribution of Senate seats for the state as a whole could be most nearly achieved without taking into account other factors, except of course the practical necessity of recognizing each county as an unbreakable unit. The Senate Committee expressed concern if certain metropolitan counties should be slightly over-represented on a population basis but it did not hesitate to recommend a plan which substantially over-represents many rural districts on such a basis.^[3] The obvious result of disregarding the substantial metropolitan fractions of Shelby, Davidson and Knox Counties was to allot the three senatorial seats thus released to more sparsely settled rural districts.

If a reasonably accurate apportionment of the 33 Senate seats throughout the state could not be accomplished except by disregarding the fractions in question, there could be no quarrel with the Senate plan adopted by the 1963 General Assembly. But consideration of proposed plans submitted by the plaintiffs in connection with the undisputed evidence in the case compels the Court to conclude that this is not the situation. These proposed apportionments for the Senate have been carefully evaluated and it is believed that the plan submitted by the plaintiffs at the last hearing, with one modification hereinafter noted, more nearly attains the goal of an accurate population distribution than any other which has been called to the attention of the Court. Under such plan Shelby and Davidson Counties are given an additional senator for each of their fractions of a ratio, resulting in only a negligible over-representation of these counties on a population basis.^[4] On the other hand, Hamilton County is not given an additional senator for its fraction of 16,139 for the evident reason that recognition of such fraction would have substantially over-represented that county.^[5] Knox County with two full ratios and a fraction of 25,159 is given two *692 senators for its two ratios. To give recognition to its fraction, another senator is given to a district composed of Knox and Anderson Counties. Such arrangement is, of course, fair and equitable so far as Knox County is concerned, but it leaves Anderson County with a voting population of only 33,554 in a district having 185,553 qualified voters. After careful consideration of suggestions to alleviate this condition, we are of the opinion that the nearest approach to an equitable solution would be to permit the voters of Anderson County to participate in the selection of all three senators. This can be accomplished by revising the plan submitted by the plaintiffs so as to provide that the 5th, 6th and 7th senatorial districts shall each comprise both Knox and Anderson Counties. As so modified the suggested plan is set forth in the appendix to this opinion.

Testimony in the record from an expert witness whose qualifications are not only impressive but unquestioned establishes that the proposed plan adheres closely to an arithmetic standard as the paramount criterion. As set forth in the stipulation of the testimony of this witness:

"In utilizing the above general procedures many different plans of apportionment were developed. Some twelve complete plans for apportioning the House and six complete plans for apportioning the Senate were developed. Each of these were examined as to the range of voting population per legislator, average deviation from the statewide average voting population per legislator, and the ratio of the largest (in voting population per legislator) to the smallest (in voting population per legislator). The plans herein submitted and now designated as 1-63H for the House and 2-63S for the Senate demonstrated the lowest range of voting population per legislator, the lowest average deviation from the statewide average voting population per legislator, and the lowest ratio of high to low with regard to voting population per legislator of all of these plans. That is to say, the plans submitted to this Court apportion the seats in their respective houses in a manner whereby the voting population per legislator throughout each house was more nearly equal than in any other plan that extensive effort and research by him has been able to devise. In the process of developing the several complete plans cited above, approximately sixty partial plans were discarded when it became apparent that the devising of districts which would deviate substantially in voting population from the statewide average would be unavoidable."

The result of this plan of apportionment for the Senate is to reduce the range of low to high voting populations under the 1963 legislation of 50,645 to 80,978 down to 59,042 voting population for the lowest in a senatorial district to 71,490 voting population for the highest in a senatorial district. It is our view that the plaintiffs' plan with the modification indicated above in conjunction with the 1963 apportionment of the House would bring about a fair, equitable and constitutional apportionment of the General Assembly of Tennessee.^[6]

*693 The defendants assert, and offer exhibits purporting to show, that the Tennessee Senate has become the second best apportioned legislative body in the United States. They point out that three-judge courts in Ohio and Illinois have upheld apportionments which depart further from a strict mathematical apportionment than does the apportionment in Tennessee under the 1963 Act, and they urgently insist that no higher standard should be applied in Tennessee than in these other states. While the argument and exhibits are indeed impressive, we do not believe that these comparisons with other states are relevant to the issue of whether the General Assembly of Tennessee has, consistent with our 1962 decision, apportioned the senate "on the basis alone of numbers of qualified voters." We do not know how many of the other states' apportionments have been tested as to constitutionality. Furthermore, every state has its own peculiar problems, such as the number, population and geographical arrangement of its subdivisions, and provisions in its own constitution. For example, Ohio's constitution recognizes fractions of ratios by giving counties with such fractions an additional representative for a specified proportional number of legislative sessions during a ten-year period. Consequently, no two states could be expected to accomplish the same degree of mathematical apportionment.

We are therefore again confronted with the question of remedy. Should the court approve a Senate plan, or should the General Assembly be given another opportunity to consider the problem before any action is taken by the Court?

While the record before us leaves no doubt that the plan submitted by the plaintiffs, as modified herein, is the best that experts have as yet been able to devise, it perhaps could never be said with finality that any plan is the ultimate and that a more representative plan could not be devised. In the 1962 proceedings, for example, the plaintiffs submitted a plan, devised after extensive study, which they believed would more nearly accomplish equal representation than would any other plan. Under that plan, the voting population per senator ranged from 59,728 to 76,000, a variance of 16,272. Further study and research, however, has resulted in the present plan under which the voting population per senator ranged from 59,042 to 71,490, a variance of only 12,478. So it may be that still better plans could be devised. Additionally, it may be that the plan set forth in the appendix could be modified by re-arranging counties in some of the rural districts upon a more acceptable basis without further departure from a voter population standard.

While the 1963 General Assembly accomplished some considerable gains in the matter of apportionment, we are persuaded that it would not be practical or feasible at this juncture for us to defer action pending possible further consideration of the problem by the General Assembly. Malapportionment in Tennessee is an evil of long standing; and the present litigation, brought to correct the situation, has been pending for more than four years. We feel, therefore, that it is not only in the public interest but in the interest of the parties as well that it should be terminated as expeditiously as possible. Moreover, this problem has already been twice examined by the General Assembly since the decision of the Supreme Court, both in regular and extraordinary sessions, and we are not convinced that further significant or substantial results could reasonably be expected through the legislative process.

The defendants in their trial brief insist that if the Court reaches the decision that there must be apportionment by court order, "it would be an abdication of its power and responsibility if it should in effect delegate to the plaintiffs *694 the legislative power to reapportion one house of the General Assembly of Tennessee to their liking and tilted in their favor."

We cannot agree that approval of a plan submitted by the plaintiffs, or for that matter by the defendants or by any other person, would in any sense be an abdication of the Court's responsibility, if such plan should comply with the standards which the Court deemed necessary to meet the requirements of the Constitution. Nevertheless, before taking any final action with reference to the plaintiffs' plan as modified in this opinion (and as so modified set forth in the appendix), we think it advisable that the defendants should be accorded an opportunity to file objections to such plan and to submit an alternative plan for consideration by the Court which they feel would equally or more nearly accomplish an apportionment on the basis of voter population.

As stated in our 1962 decision, we construed the Supreme Court's opinion to mean that "* * * we are under a positive duty to take such protective and remedial steps as will enforce the plaintiffs' rights * * *." 206 F.Supp. at p. 350. Inasmuch as the General Assembly, although making some improvements, has nevertheless failed to comply with minimum standards for constitutional legislative apportionment, our duty to intervene is unmistakable.

We, therefore, conclude that the appropriate action for the Court to take at this time is to enter an order providing as follows:

(1) The defendants will be given until February 3, 1964, to file any specific objections they may have to the Senate plan set forth in the appendix and, if they so desire, to submit an alternative plan under which Senate seats are apportioned among qualified voters of the state as equally as or more equally than they are apportioned under the said plan appearing in the appendix.

(2) If such objections and alternative plan are submitted, the Court will then pass upon any such objections and determine whether any alternative plan should be substituted in lieu of the plan set forth in the appendix, and whether either plan should be approved by the Court.

(3) Full jurisdiction will be retained by the Court and the action may be reopened at any time hereafter either on the Court's own motion or upon motion of any party.

This opinion shall constitute findings of fact and conclusions of law.

An order may be submitted in conformity with this opinion.

              PLAN FOR APPORTIONING THE SENATE OF
               THE GENERAL ASSEMBLY OF TENNESSEE
Senatorial
Districts:                           County(s)
First           -      Carter, Johnson and Washington
Second          -      Sullivan
Third           -      Cocke, Greene, Hawkins and Unicoi
Fourth          -      Claiborne, Grainger, Hamblen, Hancock,
                       Jefferson and Union
Fifth           -      Knox and Anderson
Sixth           -      Knox and Anderson
Seventh         -      Knox and Anderson
Eighth          -      Blount, Loudon and Sevier

*695
Senatorial
Districts:                           County(s)
Ninth                -       Bradley, McMinn, Meigs, Monroe and Polk
Tenth                -       Hamilton
Eleventh             -       Hamilton
Twelfth              -       Bledsoe, Cumberland, DeKalb, Rhea, Roane,
                             Sequatchie, Van Buren and White
Thirteenth           -       Campbell, Fentress, Morgan, Overton, Pickett,
                             Putnam and Scott
Fourteenth           -       Clay, Jackson, Macon, Robertson, Smith,
                             Sumner and Trousdale
Fifteenth            -       Davidson
Sixteenth            -       Davidson
Seventeenth          -       Davidson
Eighteenth           -       Davidson
Nineteenth           -       Rutherford, Williamson and Wilson
Twentieth            -       Cannon, Coffee, Franklin, Grundy, Marion
                             and Warren
Twenty-first         -       Bedford, Lincoln, Marshall, Maury and
                             Moore
Twenty-second        -       Decatur, Giles, Hardin, Hickman, Lawrence,
                             Lewis, Perry and Wayne
Twenty-third         -       Cheatham, Dickson, Houston, Humphreys,
                             Montgomery and Stewart
Twenty-fourth        -       Benton, Carroll, Gibson and Henry
Twenty-fifth         -       Crockett, Dyer, Lake, Obion and Weakley
Twenty-sixth         -       Chester, Henderson, Madison and McNairy
Twenty-seventh       -       Fayette, Hardeman, Haywood, Lauderdale
                             and Tipton
Twenty-eighth        -       Shelby
Twenty-ninth         -       Shelby
Thirtieth            -       Shelby
Thirty-first         -       Shelby
Thirty-second        -       Shelby
Thirty-third         -       Shelby

[1] For the purposes of the present opinion the metropolitan counties will be considered as Shelby having five and a fraction ratios; Davidson having three and a fraction ratios; Knox having two ratios and a fraction; and Hamilton having two ratios and a fraction. Sullivan County has one ratio and a small fraction, but no other county in the state has much as a full ratio.

[2] As used by plaintiffs, the term "ratio" means the ratio of the voting population per senator in a senatorial district to the state-wide average voting population per senator. As used in this opinion, the term "ratio" means the state-wide average voting population per senator, which is 63,420.

[3] Compare, e. g., Davidson County's disregarded fraction of 52,673 qualified voters with the following rural senatorial districts in the 1963 Act having smaller voting populations.

Senatorial District           Voting population
                              per Senator      
    No.  3                         51,714
    "    4                         52,406
    "    5                         50,887
    "   17                         52,405
    "   21                         50,943
    "   23                         51,435
    "   24                         51,487
    "   27                         52,482
    "   28                         50,645

[4] Thus under such plan Shelby County's voting population per senator is 59,922, only 3,498 less than the ratio; and Davidson County's voting population per senator is 60,733, only 2,687 less than the ratio. On the other hand, under the 1963 Act, Shelby County's voting population per senator is 71,906, or 8,486 in excess of the ratio, while Davidson County's voting population per senator is 80,978, or 17,558 in excess of the ratio.

[5] Thus Hamilton County's voting population per senator would have been 47,659, or 15,761 less than the ratio of 63,420 qualified voters.

[6] In that connection the following additional facts are informative:

The total voting population of Tennessee according to the 1960 federal census was 2,092,891. Of this number, the four metropolitan counties have 897,443 qualified voters, or 42.9% of the total. Under the 1963 Act they are given 12 senators, or 36.4% of the membership of the Senate. Under the plan set forth in the appendix, Knox shares its 3 senators with Anderson County. Accordingly, the total voting population of the four metropolitan counties, with Anderson County's 33,554 added, is 930,997, or 44.5% of the voting population of the state, and these five counties are given 15 senators, or 45.4% of the Senate, two seats short of a majority. And even if Sullivan County's one senator should be included, the five largest counties in the state, with Anderson County added, would still lack one vote in having voting control of the Senate. Thus under the 1963 House plan and the Senate plan set forth in the appendix, the four metropolitan counties, with both Sullivan and Anderson Counties added, would have control of neither House of the General Assembly.