(dissenting). I respectfully dissent from the final order of the Court. The law requires a closer adherence to the constitutional principle of "one person, one vote”.
A majority of the Court has decided to adopt a modified reapportionment plan for the Michigan Senate and Michigan House of Representatives. Today’s order does not adopt any changes in the criteria of our March 25 opinion directing the drawing of reapportionment plans under the supervision of Bernard J. Apol. In re Apportionment of State Legislature—1982, ante, p 96.
I join my colleagues in reaffirming the basic thrust of our opinion of March 25, which squarely faced the problem of an invalid apportionment scheme under our constitution and provided for the development of an apportionment plan to assure an orderly election in 1982.
However, this Court should approve a plan which would adhere to our duty to establish population as "the starting point for consideration and the controlling criterion for judgment iii legislative apportionment controversies”. Reynolds v Sims, 377 US 533, 567; 84 S Ct 1362; 12 L Ed 2d 506 (1964). Such a plan would necessitate the modification of the criteria in just one respect, the allowable population divergence. The Court had ample time to adopt an impartial plan which would not exalt district representation over people representation. 6
*215Acting under its equitable judicial power,1 the Court established in its opinion of March 25 the following primary criteria:
"Senate and House election district lines shall preserve county lines with the least cost to the federal principle of equality of population between election districts consistent with the maximum preservation of county lines and without exceeding the range of allowable divergence under the federal constitution which, until the United States Supreme Court declares otherwise, shall be deemed to be 16.4% (91.8%-108.2%).
"Where it is necessary to break county lines, because otherwise the range of allowable divergence would be exceeded, there shall be shifted the fewest cities or townships necessary to reduce the population divergence sufficiently to bring it within the range of allowable divergence.
"Because of the narrowness of the range of allowable divergence, we anticipate that only one plan will organize the counties with the least breaking of county lines.” (Italicized language added after March 25.) Ante, p 141.
The allowable population divergence of 16.4% (+8.2%) was lifted from Mahan v Howell, 410 US 315; 93 S Ct 979; 35 L Ed 2d 320 (1973). Mahan held that a plan for the Virginia House of Delegates which included an extreme allowable total deviation of 16.4% (+6.8%, —9.6%) was not unconstitutional. Written in the abstract, prima facie, such a deviation may be equitably and constitutionally acceptable. However, upon application in this state and in this case, the anticipated *216extreme divergence became commonplace. The perceived exception virtually became the norm.
A factual analysis of the Court’s plan reveals a devastating impact upon the principle of population equality. Application of the allowable population deviation in the Court’s plan results in substantial dilution or enhancement of votes in many districts throughout the state. An allowable deviation of 16.4% as applied to Michigan Senate districts which have an ideal population of 243,739 is +19,987 persons (+8.2%). This amounts to a total potential deviation from the average of almost 40,000 persons among Senate districts. The 16.4% deviation as applied to Michigan House districts with an ideal population of 84,201 is +6,904. This results in a potential variance of almost 14,000 persons among House districts.
Under the plan drawn and now approved by the Court, the total deviation between the largest and smallest Senate districts is 39,626 persons (16.26%). The largest district in terms of population, i.e., where each vote is worth the least, has a deviation exceeding the ideal by 19,788 (+8.12%); the smallest, i.e., where each vote is worth the most, is 19,838 persons ( — 8.14%) less than the ideal. A similar result is reflected in the House plan. The most populous district has a deviation of 6,870 (+8.16%); the smallest is 6,904 persons ( — 8.2%) under the ideal-sized House district. The total maximum population deviation is 13,774 (16.36%).
It becomes difficult to swallow the conclusion that a difference between Senate districts of 39,626 persons and House districts of 13,774 persons is equitable and does not impair the principle of *217equal representation. Yet, if these results truly were the exception and only occurred in the rare occasion, perhaps they would be equitably and constitutionally digestible.
What is of special significance, however, is that a great number of districts have a substantial population deviation from the ideal-sized district and fall near the extremes of the allowable divergence.
Twenty-three of 38 Senate districts have a population deviation in excess of 10,000 persons (+4.1%). Eleven of these districts fall that far below an ideal-sized district. Twelve have more than 10,000 persons above a district with perfect equality of population. Considering another comparison, 22 of 38 districts have a population deviation in excess of 12,187 (+5%). Eleven districts have a greater than 5% deviation above the ideal, and 11 fall more than 5% below the ideal. In addition, 24 Senate districts out of 38 have a deviation in excess of 9,750 persons (+4%). Finally, the average deviation from the ideal-sized district of 243,739 is 12,265 persons (+5.03%).
This is not the case where a substantial deviation from the norm is unusual. Unfortunately the plan, once unfolded, resulted in many population differentials which gravitated toward the extremes.
A similar pattern is reflected as to House districts. The average percentage variation from the ideal-sized district of 84,201 is +5.07%. Translated into persons, this reflects that there is an average deviation of 4,267 persons per House district. Fifty-four of 110 House districts have a population *218deviation in excess of 5,000 persons (+5.94%). Sixty-four of 110 have a population deviation in excess of 4,210 persons (+5%). Seventy-one have a deviation in excess of 3,368 (+4%).
It becomes apparent that the application of the Court’s criteria accentuates the extreme and not the norm. Restricting the allowable population divergence would in great measure establish population equality as the primary principle to be recognized.
The plan of the Court fails to measure up to the constitutional requirement that a court-ordered reapportionment plan should formulate districts that are "as nearly of equal population as is practicable”. In Mahan, upon analyzing the reapportionment plan for the Virginia House of Delegates, the United States Supreme Court discussed several methods of measuring population equality. One measure, the population ratio between the largest and smallest district, is the same in the Michigan plan as in Mahan, 1.18 to 1, since this plan was patterned after Mahan.
However, applying other population relationships discussed in Mahan to the Michigan plan dramatizes its constitutional deficiency. In the Michigan Senate plan, the average percentage variation from perfect equality is +5.03%. (That is 12,265 persons from the ideal-sized district). In the House plan, the average percentage deviation is +5.07%. (That is 4,269 persons from the ideal district). In Mahan, the average deviation was only ±3.89%.
Also of significance, as has been emphasized, are the number of districts which fall near the ex*219treme of the allowable deviation. In the plan of the Court, 19 of 38 Senate districts (50%) and 52 of 110 House districts (47.27%) have a population deviation greater than 6%. Only 9 out of 52 districts (17.3%) had the same deviation in Mahan. Furthermore, only 14 of 38 Michigan Senate districts (36.84%) and 39 of 110 House districts (35.45%) fall within 4% of the ideal, whereas in Mahan, the figure was significantly higher: 35 out of 52 districts (67.31%).
Thus, the surface 16.4% standard of Mahan is but the tip of the iceberg of an appropriate constitutional evaluation. While the 16.4% standard of Mahan may be met, "the fact that a 10% or 15% variation from the norm is approved in one State has little bearing on the validity of a similar variation in another State”. Swann v Adams, 385 US 440, 445; 87 S Ct 569; 17 L Ed 2d 501 (1967), quoted in Mahan v Howell, supra, 328.
An unfolding of the Court’s plan reflects that close adherence to a percentage deviation that is at the outside limits of constitutionality cannot be squared with the overriding constitutional objective of substantial equality of population among districts. The reapportionment criteria should strive to create districts as equal, not as unequal, as possible.
"Simply stated, an individual’s right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with votes of citizens living in other parts of the State.” Reynolds, supra, 568. Such a conclusion becomes more self-evident when a substantial divergence from an ideal-sized dis*220trict is present in the majority of Senate and House districts.
Today’s plan follows the Court’s opinion as closely as possible within self-imposed time constraints. Mr. Apol performed admirable service following the Court’s directives. However, rigid adherence to the criteria concerning population deviation and county lines, written in the abstract, is not required, nor is it advised, when modifications may be made to comport substantially closer to the primary principle of population equality. The Court should profess no vested interest in a certain allowable divergence when experience shows it can be substantially improved to assure constitutional acceptability and still achieve an impartial plan.
Such a modified plan could have been devised following the same boundary line criteria outlined in the reapportionment opinion, but permitting only a de minimis deviation from population equality. Cf. Gaffney v Cummings, 412 US 735; 93 S Ct 2321; 37 L Ed 2d 298 (1973); White v Regester, 412 US 755; 93 S Ct 2332; 37 L Ed 2d 314 (1973); Apportionment of Wayne County Board of Comm’rs—1982, post, p 289 (Blair Moody, Jr., J., dissenting).
A total allowable divergence of 5% (+2.5%) would certainly fall within permissible standards. Reducing the total allowable divergence to 8% (+4%) would eliminate much of the inequality extant in the present plan. Since the other criteria of the Court, particularly as interpreted by the majority, gravitates population differentials toward outside allowable limits, any further divergence would be potentially excessive. It is apparent more *221county lines would be broken than under the Court’s plan. However, population equality within acceptable limits should take precedence over unbroken county lines.
The importance of moving to a quick conclusion of this problem is recognized. However, a "rush to judgment” may not fully serve the citizens of this state. Time was available2 to develop a plan allowing only a de minimis population deviation which follows the requirements of the Equal Protection Clause — one that does not include constitutionally unacceptable line-drawing.
Reasonable adherence to boundary lines is a legitimate and useful method in an effort to minimize deliberate gerrymandering and has a genesis in Michigan history. The Court in its opinion of March 25 adopted criteria with respect to boundary lines in order to aid in the drafting of a neutral reapportionment plan.
"But if, even as a result of a clearly rational state policy of according some legislative representation to political subdivisions, population is submerged as the controlling consideration in the apportionment of seats in the particular legislative body, then the right of all of the State’s citizens to cast an effective and adequately weighted vote would be unconstitutionally impaired”. Reynolds, supra, 581. Thus, this Court cannot, under the guise of "state policy”, supersede the constitutional principle of substantial equality among election districts. Connor v Finch, 431 US 407, 420; 97 S Ct 1828; 52 L Ed 2d 465 (1977).
*222A substantial movement away from population equality in so many districts caused by excessive adherence to boundary lines turns the "constitutional requirement of equality on its head and exalt[s] district representation over people representation”. Apportionment of Wayne County, post, p 277 (Blair Moody, Jr., J., dissenting).
As was stated in Reynolds, supra, 566, "a denial of constitutionally protected rights demands judicial protection; our oath and our office require no less of us”.
For the foregoing reasons, I dissent from the adoption of the Court’s plan and would, in lieu of granting the motion for rehearing, reduce the allowable population divergence in the opinion of March 25 and direct the redrafting of a plan under such modification.
Order Denying Stay
Entered May 24, 1982. — Reporter.
On order of the Court, the ex parte application by Commissioners Kleiner, Board, Brouillette and Lurvey for stay pending United States Supreme Court appeal is considered and is hereby denied.
Justices Levin and Fitzgerald concur and say:The Democratic members of the Commission on Legislative Apportionment have filed a motion for a stay of this Court’s judgment of March 25 and order of May 21 pending their appeal to and decision by the United States Supreme Court, and have asked this Court in the interim to direct the continuation of the existing legislative districts or the adoption of one of the plans submitted by the *223Democratic members of the apportionment commission.
While this Court should ordinarily take whatever action will facilitate full exercise of the right to appeal from this Court’s decision, this Court cannot in the instant case grant a stay without immediately ordering the implementation of some alternative for this year’s election (either a further delay in the primary, an extension of the terms of legislators, a continuation of the 1972 apportioned legislative districts, an election at large, or adoption of another districting and apportionment plan) inconsistent with the decision of this Court, thereby granting, because an appeal has been taken, relief denied on the merits. A stay should, therefore, be denied.
Const 1963, art 6, § 1.
The Court was advised it would take only one week to develop such a plan.