delivered the opinion of the Court.
This case presents still another development in the efforts of the State of Florida to apportion its legislature in accordance with the requirements of the Federal Constitution. There have been previous chapters in this story. The litigation began in 1962. On June 22, 1964, in Swann v. Adams, 378 U. S. 553, we reversed the judgment of the three-judge District Court upholding the then-current legislative apportionment in Florida and remanded the case for further proceedings, consistent with the Court’s opinion in Reynolds v. Sims, 377 U. S. 533, and its companion cases. The District Court then deferred further action until the conclusion of the legislative session which convened on April 6, 1965. The *442legislature proceeded to reapportion the State on June 29, 1965. The District Court forthwith held the new plan failed to meet the requirements of the Fourteenth Amendment but approved the plan on an interim basis, limiting it to the period ending 60 days after the adjournment of the 1967 session of the Florida Legislature. This Court, finding no warrant for perpetuating what all conceded was an unconstitutional apportionment for another three years, reversed the judgment and-remanded the case to the District Court so that a valid reapportionment plan would be made effective for the 1966 elections. Swann v. Adams, 383 U. S. 210. The Florida Legislature again acted on the matter in March 1966 by adopting still another reapportionment plan which the appellants promptly attacked in the District Court.
The new plan provides for 48 senators and 117 representatives, and includes what in effect are multimember districts for each house. The senate districts range from 87,595 to 114,053 in population per senator, or from 15.09% overrepresented to 10.56% underrepresented. The ratio between the largest and the smallest district is thus 1.30 to 1. The deviation from the average population per senator is greater than 15% in one senatorial district, is greater than 14% in five more districts and is more than 10% in still six other districts. Approximately 25% of the State’s population living in one quarter of the total number of senatorial districts is underrepresented or overrepresented by at least 10%. The minimum percentage of persons that could elect a majority of 25 senators is 48.38%.
In the house the population per representative ranges from 34,584 to 48,785 or from 18.28% overrepresented to 15.27% underrepresented. The ratio between the largest and the smallest representative district is 1.41 to 1. Two districts vary from the norm by more than 18% *443and another by more than 15%, these three districts having seven of the 117 representatives. Ten other districts with 22 representatives vary from the norm by more than 10%. There is thus a deviation of more than 10% in districts which elect 29 of the 117 representatives; 24.35% of the State’s population lives in these districts. The'minimum percentage of persons that could elect a majority of 59 representatives is 47.79%.
The District Court recognized that “apportionment must be substantially on a population basis” but that “ [m] athematical exactness or precision is not required.” It went on to hold “[s]uch departures as there are from the ideal are not sufficient in number or great enough in percentages to require an upsetting of the legislative plan. . . . [WJhat deviation there is does not discriminate to any great extent against any section of the state or against either rural or urban interests.” 258 F. Supp. 819, 826, 827. Accordingly, the plan was held constitutional.
The State would have us dismiss this case for lack of standing on the part of appellants to maintain this appeal because appellants are from Dade County, Florida, which appellants concede has received constitutional treatment under the legislative plan. Appellants, however, had before the District Court their own plan which would have accorded different treatment to Dade County in some respects as compared with the legislative plan, and the alternative plan was rejected by the District Court. Moreover, the District Court has apparently consistently denied intervention to other plaintiffs, seemingly treating the appellants as representing other citizens in the State. The challenge to standing cannot succeed.
We reverse for the failure of the State to present or the District Court to articulate acceptable reasons for the variations among the populations of the various *444legislative districts with respect to both the senate and house of representatives. Reynolds v. Sims, supra, recognized that mathematical exactness is not required in state apportionment plans. De minimis deviations are unavoidable, but variations of 30% among senate districts and 40% among house districts can hardly be deemed de minimis and none of our cases suggests that differences of this magnitude will be approved, without a satisfactory explanation grounded on acceptable state policy. On the contrary, the Reynolds opinion limited the allowable deviations to those minor variations which “are based on legitimate considerations incident to the effectuation of a rational state policy.” 377 U. S. 533, 579. Thus that opinion went on to indicate that variations from a pure population standard might be justified by such state policy considerations as the integrity of political subdivisions, the maintenance of compactness and contiguity in legislative districts or the recognition of natural or historical boundary lines. Likewise, in Roman v. Sincock, 377 U. S. 695, 710, the Court stated that the Constitution permits “such minor deviations only as may occur in recognizing certain factors that are free from any taint of arbitrariness or discrimination.”
The State relies on Forty-fourth General Assembly of Colorado v. Lucas, 379 U. S. 693; Burnette v. Davis, 382 U. S. 42; and Harrison v. Schaefer, 383 U. S. 269, which were per curiam affirmances of lower court judgments in reapportionment cases. The State suggests that the plans approved in those cases involved variations in magnitude equal to or greater than those revealed by the Florida apportionment, and for that reason the judgment here should be affirmed. But in none of these cases was the issue of the validity of the differences in population between various legislative districts either raised or ruled upon in this Court. There was no occasion to explore whether or not there was ample justification for the *445challenged variations. And in Lucas v. Forty-fourth General Assembly of Colorado, 377 U. S. 713, 727, 734-735, the Court expressly reserved decision upon the validity of a variance ratio of 1.7 to 1. In any event, 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. “What is marginally permissible in one State may be unsatisfactory in another, depending upon the particular circumstances of the case.” Reynolds v. Sims, 377 U. S. 533, 578.
As this case comes to us we have no alternative but to reverse. The District Court made no attempt to explain or justify the many variations among the legislative districts. As for the State, all it suggested in either the lower court or here is that its plan comes as close as “practical” to complete population equality and that the State was attempting to follow congressional district lines. There was, however, no attempt to justify any particular deviations, even the larger ones, with respect to either of these considerations. Moreover, the State’s brief states only that the legislature followed “in most instances” the congressional boundaries, and with respect to “practicality” it seems quite obvious that the State could have come much closer to providing districts of equal population than it did. The appellants themselves placed before the court their own plan which revealed much smaller variations between the districts than did the plan approved by the District Court. Furthermore, appellants suggested to the District Court specific amendments to the legislative plan which, if they had been accepted, would have measurably reduced the population differences between many of the districts. Appellants’ own plan and their suggested amendments to the legislative plan might have been infirm in other respects but they do demonstrate that a closer approximation to equally populated districts was a feasible *446undertaking. The State, with admirable candor, states that it offered no evidence in the District Court to explain the challenged variations with respect to either the house or the senate. In its view, however, the plan should be approved on the record as it is.
We think the better view is that taken by the three-judge court in Maryland. which disapproved a legislative plan involving an overrepresentation of 14.90% and an underrepresentation of 14.38% because, as Judge Sobeloff said, there was “no showing in this case that the difference of one-third is unavoidable or justified upon any legally acceptable ground.” Maryland Citizens Committee for Fair Congressional Redistricting, Inc. v. Tawes, 253 F. Supp. 731, 733. Compare League of Nebraska Municipalities v. Marsh, 242 F. Supp. 357, disapproving a ratio of 1.6 to 1 between the smallest and the largest district absent satisfactory explanation by the State, and Paulson v. Meier, 246 F. Supp. 36, which found a ratio between the smallest and largest district of 1.39 to 1 to be unjustified on the basis offered by the State.
The appellants complain of other aspects of the plan besides unequally populated legislative districts. Under the new statute three senators were not required to run for election in 1966 but were allowed to finish their present terms expiring in 1968. These three senators, as the District Court noted, were elected in districts that are identical in territory to their districts under the legislative plan. Also, one senate and six house seats were subject to residency requirements. The District Court found no invidious discrimination in these aspects of the plan. Appellants also claim that the legislative plan discriminates invidiously by underrepresenting the populous urban counties and by overrepresenting the sparsely settled rural counties in both houses. The court below found that “what deviation there is does not *447discriminate to any great extent against any section of the state or against either rural or urban interests.” 258 F. Supp. 819, 827. In the light of our disposition of this case, however, we need not reach and decide any of these additional issues, although we note that Reynolds v. Sims indicates the constitutional impropriety of maintaining deviations from the equal population principle in deference to area and economic or other group interests. 377 U. S. 533, 579-580.
Reversed.