dissenting.

This action was brought by appellees under the Civil Rights Act, 42 U. S. C. §§ 1983, 1988, for declaratory and other relief. Their complaint alleged that New York’s congressional districting statute does not conform to the requirements of Art. I, § 2, of the United States Constitution, as those requirements are defined in Wesberry v. Sanders, 376 U. S. 1. A three-judge court was convened.

The court found, on the basis of 1960 census statistics, that the population of one of New York’s 41 congressional districts varied from the average population of the districts by 15.1%, and that 12 other districts varied from the population average by as much as 10%. It concluded that such a variation from average, without a suitable explanation, “violates constitutional requirements.” 273 F. Supp. 984, at 989. The court noted that there have been “substantial” population changes *422in these districts since 1960, and that complete accuracy in redistricting must await the results of the 1970 census, but reasoned that a suitable compromise would be to require redistricting immediately, premised on the best population figures now available. Revisions could then be made when the 1970 census statistics were released. This Court simply affirms, without elaboration or opinion.

There are, in my opinion, two principal issues here worthy of plenary consideration. First, the District Court thought it “too clear for debate” that this district-ing statute “violates constitutional requirements as enunciated by the Supreme Court.” 273 F. Supp., at 989. There are few issues in reapportionment cases that are clear beyond debate, and, with respect, the invalidity of this statute is certainly not among them. It is true that variations of as much as 18.28% were disapproved in Swann v. Adams, 385 U. S. 440, but the Court there also re-emphasized that the approval or disapproval of the variations in one State “has little bearing on the validity of a similar variation in another State.” Id., at 445. And see Reynolds v. Sims, 377 U. S. 533, 578. The impossibility of calculating the proper result in these cases from numerical variations is illustrated by Toombs v. Forison, 241 F. Supp. 65. The District Court there said specifically that “we will base any test as to the reasonableness of variances on the departure figure of 15 percent.” Id., at 70. This Court simply affirmed without opinion. 384 U. S. 210. See also Moore v. Moore, 246 F. Supp. 578, 582. Yet in this case, in which variations no greater than 15.1% are in issue, the Court again summarily affirms.

Presumably the size of the numerical variation is not alone decisive,1 but if the “particular circumstances of *423the case,” Reynolds v. Sims, supra, at 578, Swann v. Adams, supra, at 445, may be crucial to the validity of districting statutes, then surely the Court should endeavor to define what such circumstances are, and to indicate how they are relevant. Instead, the Court more and more often disposes of reapportionment cases summarily, see, e. g., Toombs v. Fortson, supra; Duddleston v. Grills, 385 U. S. 455; Kirkpatrick v. Preisler, 385 U. S. 450; Lucas v. Rhodes, ante, p. 212; and when the Court does issue an opinion, it is content simply to recite that such circumstances may be relevant, without undertaking any elucidation. See, e. g., Swann v. Adams, supra. All this has the effect of leaving the state legislatures,2 the lower courts, and even Congress3 without meaningful guidance.

*424Second, the confusion created by the Court’s reticence is compounded, in cases in which it is held that a dis-tricting statute does not satisfy constitutional requirements, by uncertainty as to the appropriate next step. See, e. g., Lucas v. Rhodes, supra. The District Court here noted that the Court’s cases provided “[l]ittle guidance or help,” but concluded, nonetheless, that immediate redistricting should be ordered, based on the best available population figures. It thought that this expedient was in the pattern set by Swann v. Adams, 383 U. S. 210. Swann, however, was merely a brief per curiam opinion, which did not purport to establish any general rule on appropriate remedies. As the elections of 1968 approach, it seems to me that the time has come for this Court to provide clearer guidance to the lower courts on the proper remedy in reapportionment cases. More particularly, some indication should be given as to what part, if any, 1960 census figures are to play, alone or in combination with later, albeit incomplete or unverified, population figures. See Lucas v. Rhodes, supra.

I would note probable jurisdiction, and set the case for argument.

The motion to affirm is granted and the judgment is affirmed.