dissenting.
Because of the uninformative nature of the Court’s reversal, some exposition of the issue in this case is necessary as a predicate for my view that the judgment of the District Court should be affirmed. My point of departure is, of course, Wesberry v. Sanders, 376 U. S. 1, a decision with which I am in continuing disagreement, see 376 U. S., at 20 et seq., 50-51, but by which I consider myself bound.
The appellants, Ohio voters, challenge the constitutionality of Ohio’s 1964 congressional redistricting statute. They assert that the redistricting plan does not satisfy the standard of population equality laid down in Wesberry v. Sanders, supra, because some of the resulting *213districts vary as much as 13% above and 18% below the population average, according to the 1960 census. In the District Court, the appellees, state officials, defended on the ground that the Ohio Legislature had properly taken into account unofficial, post-1960 population figures which were available for some counties, and which seemed to bring the 1964 redistricting into line with Wesberry.
The majority below apparently held that these unofficial population statistics were insufficient to justify the disparity among districts because they were too unreliable and not available for all areas. However, the majority went on to uphold the districting plan because
“although the varied sources of population information used by the Ohio legislature may lack uniformity of the federal census and the percentage deviation between selected Ohio districts may exceed that generally found acceptable in other states, we are unable to find that resort to the 1960 federal census in 1967 will achieve a population disparity of any lesser degree.”
Given these circumstances, I believe that the Ohio plan has not been shown to be unconstitutional, even prima facie. This Court held in Wesberry, supra, at 7-8, that “as nearly as is practicable one man’s vote in a congressional election [must] be worth as much as another’s.” However, mathematical exactness was not required of a redistricting plan, 376 U. S., at 18, and what is marginally allowable in one State may be unacceptable in another, cf. Reynolds v. Sims, 377 U. S. 533, 578. It seems to me that by failing to heed the District Court’s evident recognition that substantial shifts in population among Ohio’s congressional districts had taken place since the federal census of 1960, this Court has now given to Wesberry a Procrustean tenor which the opinion in that case does not evince.
I would affirm the judgment of the District Court.