This case was argued with Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519, which affirmed the judgment of a three-judge District Court declaring invalid Missouri's 1967 congressional districting statute. Before us ere is a judgment of a three-judge District Court for the Southern District of New York which sustained the validity of New York's 1968 congressional districting statute, N.Y.Laws 1968, c. 8, 281 F.Supp. 821 (1968). In 1967 that court had struck down an earlier districting statute apportioning New York's 41 congressional seats and had retained jurisdiction of the case pending action by the New York Legislature to redress the plan's deficiencies. The court recognized that a thorough revision of district lines might not be possible in time for the upcoming 1968 congressional election but concluded nevertheless that '(t)here are enough changes which can be superimposed on the present districts to cure the most flagrant inequalities.' 273 F.Supp. 984, 992, aff'd, 389 U.S. 421, 88 S.Ct. 578, 19 L.Ed.2d 651 (1967).

On February 28, 1968, a month and a half after the New York Legislature reconvened, the districting statute presently under attack was enacted. After a hearing, the three-judge court, on March 20, 1968, sustained the statute, stating that the districting plan afforded New York voters 'an opportunity to vote in the 1968 and 1970 elections on a basis of population equality within reasonably comparable districts.' 281 F.Supp., at 826. We noted probable jurisdiction. 393 U.S. 819, 89 S.Ct. 115, 21 L.Ed.2d 91 (1968). We reverse insofar as the judgment of the District Court sustains the plan for use in the 1970 congressional election.

Appellant levels two constitutional attacks against the statute: (1) that the statute violates the equal-population principle of Wesberry v. Sanders, 376 U.S. 1, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964), and (2) that the statute represents a systematic and intentional partisan gerrymander violating Art. I, § 2, of the Constitution and the Fourteenth Amendment. We do not reach, and intimate no view upon the merits of, the attack upon the statute as a constitutionally impermissible gerrymander. We hold that reversal of the District Court's judgment is compelled by our decision today in Kirkpatrick v. Preisler, supra, which elucidates the command of Wesberry that congressional districting meet the standard of equal representation for equal numbers of people as nearly as is practicable.

The District Court correctly held in its 1967 opinion that 'there is a burden on the proponent of any districting plan to justify deviations from equality.' 273 F.Supp., at 987. The District Court took no testimony on the question of justification at the hearing held to consider the 1968 statute. Recognizing that the statute, which was enacted with virtually no debate on its merits in either house of the New York Legislature, was the work of a Joint Legislative Committee, the court's 1968 opinion refers to the Report of the Joint Committee as the source of the justifications relied upon as sufficient to sustain the population disparities created by the plan. 281 F.Supp., at 823—824. We have been referred to the same source.

The Report recites that the Committee 'gave priority to the population totals in the several districts' as they appeared in the 1960 decennial census and that 'very limited consideration was given to population shifts within the State since 1960. The Report recites further that '(o)ther considerations were the geographical conformation of the area to be districted, the maintenance of county integrity, the facility by which the various Boards of Elections can 'tool up' for the forthcoming (1968) primary election, equality of population within the region, and equality of population throughout the state.' Interim Report of the Joint Legislative Committee on Reapportionment of N.Y. State Legislature (1968).

The heart of the scheme, however, lay in the decision to treat seven sections of the State as homogeneous regions and to divide each region into congressional districts of virtually identical population. Thirty-one of New York's 41 congressional districts were constructed on that principle. The remaining 10 istricts were composed of groupings of whole counties. A chart showing the population of each district under the 1968 statute appears in the Appendix to this opinion. The seven regions are: (a) Suffolk and Nassau Counties on Long Island with five districts having an average population of 393,391 and a maximum deviation from that average of 208; (b) Queens County with four districts having an average population of 434,672 and a maximum deviation from that average of 120; (c) Kings County plus a district made up of part of Kings and part of Queens, and a district made up of Richmond County and part of Kings, with seven districts having an average population of 417,171 and a maximum deviation from that average of 307; (d) New York and Bronx Counties with eight districts having an average population of 390,415 and a maximum deviation from that average of 496; (e) Westchester and Putnam Counties with two districts having an average population of 420,307 and a maximum deviation from that average of 161; (f) Wayne plus part of Monroe and the remainder of Monroe plus four other counties with two districts having an average population of 410,688 and a maximum deviation from that average of 256; and (g) Erie and Niagara Counties with three districts having an average population of 435,652 and a maximum deviation from that average of 228. The 10 remaining 'North country' districts were composed of groupings of whole counties.

It is clear that our decision in Kirkpatrick v. Preisler, supra, compels the conclusion that this scheme is unconstitutional. We there held, 394 U.S., at 531, 89 S.Ct., at 1239, that 'the command of Art. I, § 2, that States create congressional districts which provide equal representation for equal numbers of people permits only the limited population variances which are unavoidable despite a good-faith effort to achieve absolute equality, or for which justification is shown.' The general command, of course, is to equalize population in all the districts of the State and is not satisfied by equalizing population only within defined sub-states. New York could not and does not claim that the legislature made a good-faith effort to achieve precise mathematical equality among its 41 congressional districts. Rather, New York tries to justify its scheme of constructing equal districts only within each of seven sub-states as a means to keep regions with distinct interests intact. But we made clear in Kirkpatrick that 'to accept population variances, large or small, in order to create districts with specific interest orientations is antithetical to the basic premise of the constitutional command to provide equal representation for equal numbers of people.' To accept a scheme such as New York's would permit groups of districts with defined interest orientations to be overrepresented at the expense of districts with different interest orientations. Equality of population among districts in a sub-state is not a justification for inequality among all the districts in the State.

Nor are the variations in the 'North country' districts justified by the fact that these districts are constructed of entire counties. Kirkpatrick v. Preisler, supra.

We appreciate that the decision of the District Court did not rest entirely on an appraisal of the merits of the New York plan. As noted earlier, when the three-judge District Court in 1967 held the then-existing districting plan unconstitutional, it recognized that the imminence of the 1968 election made redistricting an unrealistic possibility and therefore said only that '(t)here are enough changes which can be superimposed on the present districts to cure the most flagrant inequalities.' 273 F.Supp., at 992. On February 26, 1968, the New York Legislature enacted the plan before us. On March 20, 1968, the District Court approved the plan for both the 1968 and 1970 congressional elections. Since the 1968 primary election was only three months away on March 20, we cannot say that there was error in p rmitting the 1968 election to proceed under the plan despite its constitutional infirmities. See Kilgarlin v. Hill, 386 U.S. 120, 121, 87 S.Ct. 820, 821, 17 L.Ed.2d 771 (1967); Martin v. Bush, 376 U.S. 222, 223, 84 S.Ct. 709, 11 L.Ed.2d 656 (1964); Kirkpatrick v. Preisler, 390 U.S. 939, 88 S.Ct. 1053, 19 L.Ed.2d 1129 (1968). But ample time remains to promulgate a plan meeting constitutional standards before the election machinery must be set in motion for the 1970 election. We therefore reverse the judgment of the District Court insofar as it approved the plan for use in the 1970 election and remand the case for the entry of a new judgment consistent with this opinion.

It is so ordered.

Judgment of District Court reversed and case remanded.

APPENDIX.

POPULATION OF NEW YORK'S CONGRESSIONAL DISTRICTS UNDER 1968 PLAN.

C.D. Dev. % Description.

1 393,585 - 3.845 Part of Suffolk.

2 393,465 - 3.874 Part of Suffolk, Part of Nassau.

3 393,434 - 3.882 Part of Nassau.

4 393,183 - 3.943 Part of Nassau.

5 393,288 - 3.918 Part of Nassau.

6 434,615 + 6.178 Part of Queens.

7 434,750 + 6.212 Part of Queens.

8 434,552 + 6.163 Part of Queens.

9 434,770 + 6.217 Part of Queens.

10 417,122 + 1.905 Part of Queens, Part of Kings.

11 417,090 + 1.897 Part of Kings.

12 417,298 + 1.948 Part of Kings.

13 417,040 + 1.885 Part of Kings.

14 417,080 + 1.895 Part of Kings.

15 417,090 + 1.898 Part of Kings.

16 417,478 + 1.992 Richmond, Part of Kings.

17 390,742 - 4.540 Part of New York.

18 390,861 - 4.511 Part of New York.

19 390,023 - 4.715 Part of New York.

20 390,363 - 4.632 Part of New York.

21 390,552 - 4.586 Part of New York, Part of Bronx.

22 390,492 - 4.601 Part of Bronx.

23 390,228 - 4.665 Part of Bronx.

24 390,057 - 4.707 Part of Bronx.

25 420,146 + 2.644 Putnam, Part of Westchester.

26 420,467 + 2.722 Part of Westchester.

27 409,349 Rockland, Orange, Sullivan, Delaware.

28 396,122 - 3.225 Dutchess, Ulster, Columbia, Greene,

29 425,822 + 4.031 Albany, Schenectady.

30 415,030 + 1.394 Rensselaer, Saratoga, Washington,

31 425,905 + 4.051 Clinton, St. Lawrence, Jefferson,

32 385,406 - 5.843 Oneida, Madison, Herkimer.

33 415,333 + 1.468 Chemung, Broome, Tioga, Tompkins.

34 423,028 + 3.348 Onondaga.

35 386,148 - 5.662 Ontario, Yates, Seneca, Cayuga,

36 410,943 + 0.396 Part of Monroe, Wayne.

37 410,432 + 0.271 Part of Monroe, Orleans, Genesee,

38 382,277 - 6.608 Chautaqua, Cattaraugus, Allegany,

39 435,393 + 6.369 Part of Erie.

40 435,684 + 6.440 Part of Erie, Niagara.

41 435,880 + 6.488 Part of Erie.

State Mean............................ 409,324

Largest District (41st C. D.)......... 435,880

Smallest District (38th C. D.)........ 382,277

Citizen Population Variance (largest district population

divided by the smallest district population). 1.139 to 1

Maximum Deviation above State Mean..... 6.488%

Maximum Deviation below State Mean..... 6.608%

Mr. Justice FORTAS, concurring.

I concur in the judgment of the Court and in its opinion except to the extent that the opinion relies upon the Court's opinion in the Missouri redistricting cases, Kirkpatrick v. Preisler, 394 U.S. 526, 89 S.Ct. 1225, 22 L.Ed.2d 519, which I have not joined for the reasons stated in my concurring opinion in those cases.

New York does not attempt to defend its plan as a good-faith effort to achieve districts of approximate equality. It argues that it devised a plan based upon the grouping of districts into regions. I agree with the majority that, for purposes of the congressional districting here involved, the State may not substantially or grossly disregard population or residence figures in order to recognize regional groupings within the State. See my dissent in Avery v. Midland County, 90 U.S. 474, 495, 88 S.Ct. 1114, 1126, 20

Mr. Justice HARLAN, with whom Mr. Justice STEWART joins, dissenting.

Whatever room remained under this Court's prior decisions for the free play of the political process in matters of reapportionment is now all but eliminated by today's Draconian judgments. Marching to the nonexistent 'command of Art. I, § 2' of the Constitution,1 the Court now transforms a political slogan into a constitutional absolute. Straight indeed is the path of the righteous legislator. Slide rule in hand, he must avoid all thought of county lines, local traditions, politics, history, and economics, so as to achieve the magic formula: one man, one vote.

As my Brothers WHITE and FORTAS demonstrate, insistence on mathematical perfection does not make sense even on its own terms. Census figures themselves are inexact; our mobile population rapidly renders them obsolete; large groups of ineligible voters are unevenly distributed throughout the State. Nevertheless, the Court refuses to permit any room for legislative common sense to compensate for Census Bureau inadequacies. If no 'scientific' data are available to justify a divergence from census figures, the Court holds that nothing can be done—'we mean to open no avenue for subterfuge.' Kirkpatrick v. Preisler, 394 U.S., at 535, 89 S.Ct., at 1231.

This all-pervasive distrust of the legislative process is completely alien to established notions of judicial review. See Butler v. Pennsylvania, 10 How. 402, 13 L.Ed. 472 (1851); Davis v. Department of Labor, 317 U.S. 249, 63 S.Ct. 225, 87 L.Ed. 246 (1942); Flemming v. Nestor, 363 U.S. 603, 80 S.Ct. 1367, 4 L.Ed.2d 1435 (1960). Nor does it have precedent in the prior reapportionment decisions themselves. 'Reynolds v. Sims (377 U.S. 553, 84 S.Ct. 1362, 12 L.Ed.2d 506) recognized that mathematical exactness is not required in state apportionment plans. De minimis deviations are unavoidable * * *.' Swann v. Adams, 385 U.S. 440, 444, 87 S.Ct. 569, 17 L.Ed.2d 501 (1967); see also Wesberry v. Sanders, 376 U.S. 1, 18, 84 S.Ct. 526, 535, 11 L.Ed.2d 481 (1964).2

Even more important, the Court's exclusive concentration upon arithmetic blinds it to the realities of the political process, as the Rockefeller case makes so clear. The fact of the matter is that the rule of absolute equality is perfectly compatible with 'gerrymandering' of the worst sort. A computer may grind out district lines which can totally frustrate the popular will on an overwhelming number of critical issues. The legislature must do more than satisfy one man, one vote; it must create a structure which will in fact as well as theory be responsive to the sentiments of the community On the record before us, however, there is absolutely no indication that the New York Legislature can satisfy this Court's demand for absolute equality and yet create a structure which will permit New York's multitude of political groups to have a fair chance at having their voices heard in Congress.

Even the appellant himself does not suggest that it is possible to create a proper apportionment plan which is at the same time consistent with the demands of perfect mathematical equality. The plan he advances contemplates a maximum deviation of 4.7% from the state average, which represents an improvement of only 1.9 percentage points on the State's 6.6% deviation. Moreover, under the State's plan, a majority of the congressional delegation can represent no less than 49.3% of the population. The appellant's scheme 'improves' this figure by 0.5%, increasing the number to 49.8% See Appellant's Appendix D. Perfection, however, is still 0.2% away.

Although the appellant's plan offers such marginal benefits of voting egalitarianism, and although the record contains no suggestion of any other plan which even arguably permits the coherent expression of the popular will, the Court rejects the legislature's considered proposal simply because it seeks to remain true to traditional county and regional lines. In doing so, the majority ignores the salutary warning to be found in Reynolds v. Sims, 377 U.S. 533, 578—579, 84 S.Ct. 1362, 1390, 12 L.Ed.2d 506 (1964): 'Indiscriminate districting, without any regard for political subdivision or natural or historical boundary lines, may be little more than an open invitation to partisan gerrymandering.' Yet, today the Court condemns the legislature's approach because it 'permit(s) groups of districts with defined interest orientations to be over represented at the expense of districts with different interest orientations.' Wells v. Rockefeller, 394 U.S., at 546, 89 S.Ct., at 1237. Of course, all districting decisions inevitably involve choices between different interest groups. But as Reynolds recognized, legislatures prefer to follow traditional county and regional lines so that the demands of blatant partisanship will be tempered by the constraints of history and tradition. If the Court believes it has struck a blow today for fully responsive representative democracy, it is sorely mistaken. Even more than in the past, district lines are likely to be drawn to maximize the political advantage of the party temporarily dominant in public affairs.

We do not deal here with the hopelessly malapportioned legislature unwilling to set its own house in order. Rather, the question before us is whether the Constitution requires that mathematics be a substitute for common sense in the art of statecraft. As I do not think that the apportionment plans submitted by the States of New York and Missouri can properly be regarded as offensive to the requirement of equality imposed in Wesberry—a case whose constitutional reasoning I still find it impossible to swallow, but by whose dictate I consider myself bound—I dissent.

I would reverse the judgments of the District Court in the Missouri cases and affirm the decision of the District Court in the New York case.L.Ed.2d 45 (1968).

Mr. Justice WHITE, dissenting.

I have consistently joined the Court's opinions which establish as one of the ground rules for legislative districting that single member districts should be substantially equal in population. I would not now dissent if the Court's present judgments represented a measurable contribution to the ends whic I had thought the Court was pursuing in this area, or even if I thought the opinions not very useful but not harmful either. With all due respect, however, I am firmly convinced that the Court's new rulings are unduly rigid and unwarranted applications of the Equal Protection Clause which will unnecessarily involve the courts in the abrasive task of drawing district lines.

Accepting for constitutional purposes that a State may assign the task of apportioning its legislature or congressional delegation to the legislature itself, I would not quibble with the legislative judgment if variations between districts were acceptably small. And I would be willing to establish a population variation figure which if not exceeded would normally not call for judicial intervention. As a rule of thumb, a variation between the largest and the smallest district of no more than 10% to 15% would satisfy me, absent quite unusual circumstances not present in any of these cases. At the very least, at this trivial level, I would be willing to view state explanations of the variance with a more tolerant eye.

This would be far more reasonable than the Court's demand for an absolute but illusory equality or for an apportionment plan which approaches this goal so nearly that no other plan can be suggested which would come nearer. As Mr. Justice FORTAS demonstrates, the 1960 census figures were far from accurate when they were compiled by professional enumerators and statisticians bent on precision, in 1960. Massive growth and shifts in population since 1960 made the 1960 figures even more inaccurate by 1967. That is why a new census is taken every 10 years. When the Court finds a 3% variation from substantially inexact figures constitutionally impermissible it is losing perspective and sticking at a trifle.

It also seems arbitrary for the majority to discard the suggestion of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), that if a legislature seeks an apportionment plan which respects the boundaries of political subdivisions, some variations from absolute equality would be constitutionally permissible. Of course, Reynolds involved state legislative apportionment and took pains to say that there may be more leeway in that context. But the Court invokes Reynolds today and in no way distinguishes federal from state districting.

Reynolds noted that '(i)ndiscriminate districting, without any regard for political subdivision or natural or historical boundary lines, may be little more than an open invitation to partisan gerrymandering.' 377 U.S., at 578—579, 84 S.Ct., at 1390. The Court nevertheless now rules that regard for these boundaries is no justification for districts which vary no more than 3% from the norm where another plan which may have no regard for district lines reduces the variation to an even smaller figure. I have similar objections to the Court's rejection of geographical compactness as an acceptable justification for minor variations among congressional districts. This rejection of the virtues of compactness will not be lost on those who would use congressional and legislative districting to bury their political opposition.

In reality, of course, districting is itself a gerrymandering in the sense that it represents a complex blend of political, economic, regional, and historical considerations. In terms of the gerrymander, the situation will not be much different if equality means what it literally says—a zero variation—rather than only 'substantial' equality which would countenance some variations among legislative districts. Either standard will prevent a minority of the population or a minority party from consistently controlling the state legislature or a congressional delegation, and both are powerful forces toward equalizing voter influence on legislative performance. In terms of effective representation for all voters there are only minuscule differences between the two standards. But neither rule can alone prevent de iberate partisan gerrymandering if that is considered an evil which the Fourteenth Amendment should attempt to proscribe.

Today's decisions on the one hand require precise adherence to admittedly inexact census figures, and on the other downgrade a restraint on a far greater potential threat to equality of representation, the gerrymander. Legislatures intent on minimizing the representation of selected political or racial groups are invited to ignore political boundaries and compact districts so long as they adhere to population equality among districts using standards which we know and they know are sometimes quite incorrect. I see little merit in such a confusion of priorities.

Moreover, today's decisions will lead to an unnecessary intrusion of the judiciary into legislative business. It would be one thing if absolute equality were possible. But, admittedly, it is not. The Court may be groping for a cleancut, per se rule which will minimize confrontations between courts and legislatures while also satisfying the Fourteenth Amendment. If so, the Court is wide of the mark. Today's results simply shift the area of dispute a few percentage points down the scale; the courts will now be engaged in quibbling disputes over such questions as whether a plan with a 1% variation is 'better' than one with a larger variation, say 1.1% or even 2%. If county and municipal boundaries are to be ignored, a computer can produce countless plans for absolute population equality, one differing very little from another, but each having its own very different political ramifications. Ultimately, the courts may be asked to decide whether some families in an apartment house should vote in one district and some in another, if that would come closer to the standard of apparent equality. Using the spacious language of the Equal Protection Clause to inject the courts into these minor squabbles is an unacceptable pre-emption of the legislative function. Not only will the Court's new rule necessarily precipitate a new round of congressional and legislative districting, but also I fear that in the long run the courts, rather than the legislatures or nonpartisan commissions, will be making most of the districting decisions in the several States. Since even at best, with compact and equal districts, the final boundary lines unavoidably have significant political repercussions, the courts should not draw district lines themselves unnecessarily. I therefore dissent.